Datasets:

Udbhav44

/

qu-bit

like 0

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'query': 'Perform a comprehensive update on custom field options with id 789 and get details for a set of issues within a Forge app.', 'solutions': [{'tool': 'issue_custom_field_options', 'arguments': [{'argument_name': 'customFieldOption/:id', 'argument_value': '789'}]}, {'tool': 'issue_custom_field_options_(apps)', 'arguments': [{'argument_name': 'field/:fieldKey/option', 'argument_value': '<option_details>'}]}, {'tool': 'issue_custom_field_values_(apps)', 'arguments': [{'argument_name': 'app/field/value', 'argument_value': 'Forge'}]}]}, {'query': "Update the name and description of a field configuration with fieldConfigId 456 used for managing issue fields in Jira, and then add 'Epic' and 'Task' issue types to a custom field context with fieldId ABC1234 and customId ABCD and for better project management.", 'solutions': [{'tool': 'issue_field_configurations', 'arguments': [{'argument_name': 'fieldConfigId', 'argument_value': '456'}, {'argument_name': 'updated_details', 'argument_value': {'name': '<new_name>', 'description': '<new_description>'}}]}, {'tool': 'issue_custom_field_contexts', 'arguments': [{'argument_name': 'fieldId', 'argument_value': 'ABC1234'}, {'argument_name': 'contextId', 'argument_value': 'ABCD'}, {'argument_name': 'issuetype', 'argument_value': ['Epic', 'Task']}]}]}, {'query': 'Retrieve all remote issue links for the issue ABCD1234, obtain a list of all issue resolution values, and search for issues using JQL.', 'solutions': [{'tool_name': 'issue_remote_links', 'arguments': [{'argument_name': 'issue/:issueIdOrKey/remotelink', 'argument_value': 'ABCD1234'}]}, {'tool_name': 'issue_resolutions', 'arguments': [{'argument_name': 'resolution', 'argument_value': 'resolution_values'}]}, {'tool_name': 'issue_search', 'arguments': [{'argument_name': 'search', 'argument_value': 'JQL'}]}]}] Current Query : Perform a comprehensive update on custom field options with id 789 and get details for a set of issues within a Forge app. Current Solution : { "APIs": [ { "tool": "issue_custom_field_options", "arguments": [ { "argument_name": "customFieldOption/:id", "argument_value": "789" } ] }, { "tool": "issue_custom_field_options_(apps)", "arguments": [ { "argument_name": "field/:fieldKey/option", "argument_value": "<option_details>" } ] }, { "tool": "issue_custom_field_values_(apps)", "arguments": [ { "argument_name": "app/field/value", "argument_value": "Forge" } ] } ] } Answer: [/INST] { "APIs": [ { "tool": "issue_custom_field_options", "arguments": [ { "argument_name": "customFieldOption/:id", "argument_value": "789" } ] }, { "tool": "issue_custom_field_options_(apps)", "arguments": [ { "argument_name": "field/:fieldKey/option", "argument_value": "<option_details>" } ] }, { "tool": "issue_custom_field_values_(apps)", "arguments": [ { "argument_name": "app/field/value", "argument_value": "Forge" } ] } ] } </s>

<s>[INST] Given a query, solution and available APIs, optimize the number of APIs calls in the solution. Do not use API calls unless absolutely necessary. Look for redundancy and check for mistakes in the combination of API calls. Return current solution if no changes are necessary. Current Query : Get all work items similar to TKT-456, summarize them, create issues from that summary, and prioritize them Current Solution : [ { "tool_name":"get_similar_work_items", "arguments":[ { "argument_name": "work_id", "argument_value": "TKT-456" } ] }, { "tool_name": "create_actionable_tasks_from_text", "arguments": [ { "argument_name": "text", "argument_value": "$$PREV[0]" } ] }, { "tool_name": "prioritize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$$PREV[1]" } ] } ] } ] Available APIs : [{'tool': 'get_similar_work_items', 'description': 'Returns a list of work items that are similar to the given work item', 'arguments': [{'argument_name': 'work_id', 'argument_description': 'The ID of the work item for which you want to find similar items', 'argument_type': 'string', 'examples': []}]}, {'tool': 'works_list', 'description': 'Returns a list of work items matching the request', 'arguments': [{'argument_name': ' applies_to_part', 'argument_description': 'Filters for work belonging to any of the provided parts', 'argument_type': 'array of strings', 'examples': ['FEAT-123', 'ENH-123', 'PROD-123', 'CAPL-123']}, {'argument_name': 'created_by', 'argument_description': 'Filters for work created by any of these users', 'argument_type': 'array of strings', 'examples': ['DEVU-123']}, {'argument_name': 'issue.priority', 'argument_description': 'Filters for issues with any of the provided priorities. Allowed values: p0, p1, p2,p3', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'issue.rev_orgs', 'argument_description': 'Filters for issues with any of the provided Rev organizations', 'argument_type': 'array of strings', 'examples': ['REV-123']}, {'argument_name': 'limit', 'argument_description': "The maximum number of works to return. The default is '50'", 'argument_type': 'integer (int32)', 'examples': []}, {'argument_name': 'owned_by', 'argument_description': 'Filters for work owned by any of these users', 'argument_type': 'array of strings', 'examples': ['DEVU-123']}, {'argument_name': 'stage.name', 'argument_description': 'Filters for records in the provided stage(s) by name', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'ticket.needs_response', 'argument_description': 'Filters for tickets that need a response', 'argument_type': 'boolean', 'examples': []}, {'argument_name': 'ticket.rev_org', 'argument_description': 'Filters for tickets associated with any of the provided Rev organizations', 'argument_type': 'array of strings', 'examples': ['REV-123']}, {'argument_name': 'ticket.severity', 'argument_description': 'Filters for tickets with any of the provided severities. Allowed values:blocker,high, low, medium', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'ticket.source_channel', 'argument_description': 'Filters for tickets with any of the provided source channels', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'ticket.source_channel', 'argument_description': 'Filters for tickets with any of the provided source channels', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'type', 'argument_description': 'Filters for work of the provided types. Allowed values issue, ticket, task', 'argument_type': 'array of strings', 'examples': []}]}, {'tool': 'add_work_items_to_sprint', 'description': 'Adds the given work items to the sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_description': 'A list of work item IDs to be added to the sprint.', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'sprint_id', 'argument_description': 'The ID of the sprint to which the work items should be added', 'argument_type': 'str', 'examples': []}]}, {'tool': 'summarize_objects', 'description': 'Summarizes a list of objects. The logic of how to summarize a particular object type is an internal implementation detail.', 'arguments': [{'argument_name': 'objects', 'argument_description': 'List of objects to summarize', 'argument_type': 'array of objects', 'examples': []}]}] Final solution (same format as current solution): [/INST] [ { "tool_name":"get_similar_work_items", "arguments":[ { "argument_name": "work_id", "argument_value": "TKT-456" } ] }, { "tool_name": "create_actionable_tasks_from_text", "arguments": [ { "argument_name": "text", "argument_value": "$$PREV[0]" } ] }, { "tool_name": "prioritize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$$PREV[1]" } ] } ] } ] </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [[{'query': 'Create a new part, retrieve its information, and list parts created by DEVU-12345.', 'solutions': [{'tool': 'parts.create', 'arguments': []}, {'tool': 'parts.get', 'arguments': [{'argument_name': 'id', 'argument_value': '$$PREV[0]'}]}, {'tool': 'parts.list', 'arguments': [{'argument_name': 'created_by', 'argument_value': ['DEVU-12345']}]}]}], [{'query': 'Create a new part, update its details, and fetch the list of parts owned by DEVU-45678.', 'solutions': [{'tool': 'parts.create', 'arguments': []}, {'tool': 'parts.update', 'arguments': []}, {'tool': 'parts.list', 'arguments': [{'argument_name': 'owned_by', 'argument_value': ['DEVU-45678']}]}]}], [{'query': 'Create a new part, retrieve its information, update its details, delete a different part, and list parts created by DEVU-78901.', 'solutions': [{'tool': 'parts.create', 'arguments': []}, {'tool': 'parts.get', 'arguments': [{'argument_name': 'id', 'argument_value': '$$PREV[0]'}]}, {'tool': 'parts.update', 'arguments': []}, {'tool': 'parts.delete', 'arguments': [{'argument_name': 'id', 'argument_value': '$$PREV[2]'}]}, {'tool': 'parts.list', 'arguments': [{'argument_name': 'created_by', 'argument_value': ['DEVU-78901']}]}]}]] Current Query : Create a new part, retrieve its information, and list parts created by DEVU-12345. Current Solution : { "APIs": [ { "tool": "parts.create", "arguments": [] }, { "tool": "parts.get", "arguments": [ { "argument_name": "id", "argument_value": "$$PREV[0]" } ] }, { "tool": "parts.list", "arguments": [ { "argument_name": "created_by", "argument_value": ["DEVU-12345"] } ] } ] } Answer: [/INST] { "APIs": [ { "tool": "parts.create", "arguments": [] }, { "tool": "parts.get", "arguments": [ { "argument_name": "id", "argument_value": "$$PREV[0]" } ] }, { "tool": "parts.list", "arguments": [ { "argument_name": "created_by", "argument_value": ["DEVU-12345"] } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'Query': 'What are my all issues in the triage stage under part FEAT-123? Summarize them.', 'Solution': [{'tool_name': 'who_am_i', 'arguments': []}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'stage.name', 'argument_value': 'triage'}, {'argument_name': 'applies_to_part', 'argument_value': 'FEAT-123'}, {'argument_name': 'owned_by', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'Summarize high severity tickets from the customer UltimateCustomer', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'UltimateCustomer'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'Prioritize my P0 issues and add them to the current sprint', 'Solution': [{'tool_name': 'who_am_i', 'arguments': []}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'issue.priority', 'argument_value': 'p0'}, {'argument_name': 'owned_by', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}, {'tool_name': 'get_sprint_id', 'arguments': []}, {'tool_name': 'add_work_items_to_sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_value': '$$PREV[2]'}, {'argument_name': 'sprint_id', 'argument_value': '$$PREV[3]'}]}]}] Current Query : What are my all issues in the triage stage under part FEAT-123? Summarize them. Current Solution : { "tools": [ { "tool": "who_am_i", "arguments": [] }, { "tool": "works_list", "arguments": [ { "argument_name": "stage.name", "argument_value": "triage" }, { "argument_name": "applies_to_part", "argument_value": "FEAT-123" }, { "argument_name": "owned_by", "argument_value": "$$PREV[0]" } ] }, { "tool": "summarize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$$PREV[1]" } ] } ] } Answer: [/INST] [ { "tool_name": "whoami", "arguments": [] }, { "tool_name": "works_list", "arguments": [ { "argument_name": "stage.name","argument_value": ["triage"] }, { "argument_name": "applies_to_part", "argument_value": ["FEAT-123"] }, { "argument_name": "owned_by", "argument_value": ["$$PREV[0]"] }, { "argument_name": "type", "argument_value": ["issue"] } ] }, { "tool_name": "summarize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$$PREV[1]" } ] } ] </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [[{'query': 'Create a new part, retrieve its information, and list parts created by DEVU-12345.', 'solutions': [{'tool': 'parts.create', 'arguments': []}, {'tool': 'parts.get', 'arguments': [{'argument_name': 'id', 'argument_value': '$$PREV[0]'}]}, {'tool': 'parts.list', 'arguments': [{'argument_name': 'created_by', 'argument_value': ['DEVU-12345']}]}]}], [{'query': 'Create a new part, update its details, and fetch the list of parts owned by DEVU-45678.', 'solutions': [{'tool': 'parts.create', 'arguments': []}, {'tool': 'parts.update', 'arguments': []}, {'tool': 'parts.list', 'arguments': [{'argument_name': 'owned_by', 'argument_value': ['DEVU-45678']}]}]}], [{'query': 'Create a new part, retrieve its information, update its details, delete a different part, and list parts created by DEVU-78901.', 'solutions': [{'tool': 'parts.create', 'arguments': []}, {'tool': 'parts.get', 'arguments': [{'argument_name': 'id', 'argument_value': '$$PREV[0]'}]}, {'tool': 'parts.update', 'arguments': []}, {'tool': 'parts.delete', 'arguments': [{'argument_name': 'id', 'argument_value': '$$PREV[2]'}]}, {'tool': 'parts.list', 'arguments': [{'argument_name': 'created_by', 'argument_value': ['DEVU-78901']}]}]}]] Current Query : Create a new part, retrieve its information, and list parts created by DEVU-12345. Current Solution : { "APIs": [ { "tool": "parts.create", "arguments": [] }, { "tool": "parts.get", "arguments": [ { "argument_name": "id", "argument_value": "$PREV[0]" } ] }, { "tool": "parts.list", "arguments": [ { "argument_name": "created_by", "argument_value": ["DEVU-12345"] } ] } ] } Answer: [/INST] { "APIs": [ { "tool": "parts.create", "arguments": [] }, { "tool": "parts.get", "arguments": [ { "argument_name": "id", "argument_value": "$$PREV[0]" } ] }, { "tool": "parts.list", "arguments": [ { "argument_name": "created_by", "argument_value": ["DEVU-12345"] } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'query': 'Retrieve the current announcement banner configuration including required permissions for access, and then get details of all application roles in Jira.', 'solutions': [{'tool': 'announcement_banner', 'arguments': [{'argument_name': 'include_permissions', 'argument_value': 'true'}]}, {'tool': 'application_roles', 'arguments': [{'argument_name': 'applicationrole', 'argument_value': 'Jira'}]}]}, {'query': "Retrieve the metadata for an attachment with ID '12345,' and then create a new custom field named 'Release Version' for tracking software releases in Jira.", 'solutions': [{'tool': 'issue_attachments', 'arguments': [{'argument_name': 'attachment_id', 'argument_value': '12345'}]}, {'tool': 'issue_fields', 'arguments': [{'argument_name': 'field_details', 'argument_value': {'name': 'Release Version'}}]}]}, {'query': 'Retrieve default columns for the issue navigator, manage notification schemes for scheme id LM10(get existing, create new, update, and customize), and update default columns if necessary.', 'solution': [{'tool_name': 'issue_navigator_settings', 'arguments': [{'argument_name': 'get_default_columns', 'argument_value': ''}]}, {'tool_name': 'issue_notification_schemes', 'arguments': [{'argument_name': 'create_notification_scheme', 'argument_value': {'scheme_details': 'Provide details for the new notification scheme.'}}, {'argument_name': 'update_notification_scheme', 'argument_value': {'scheme_id': 'LM10'}}, {'argument_name': 'add_notifications_to_scheme', 'argument_value': {'scheme_id': 'LM10'}}]}]}] Current Query : Retrieve the current announcement banner configuration including required permissions for access, and then get details of all application roles in Jira. Current Solution : { "APIs": [ { "tool": "announcement_banner", "arguments": [ { "argument_name": "include_permissions", "argument_value": "true" } ] }, { "tool": "application_roles", "arguments": [ { "argument_name": "applicationrole", "argument_value": "Jira" } ] } ] } Answer: [/INST] { "APIs": [ { "tool": "announcement_banner", "arguments": [ { "argument_name": "include_permissions", "argument_value": "true" } ] }, { "tool": "application_roles", "arguments": [ { "argument_name": "applicationrole", "argument_value": "Jira" } ] } ] } </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Get all work items similar to TKT-456, summarize them, create issues from that summary, and prioritize them APIs: [{'tool': 'get_similar_work_items', 'description': 'Returns a list of work items that are similar to the given work item', 'arguments': [{'argument_name': 'work_id', 'argument_description': 'The ID of the work item for which you want to find similar items', 'argument_type': 'string', 'examples': []}]}, {'tool': 'works_list', 'description': 'Returns a list of work items matching the request', 'arguments': [{'argument_name': ' applies_to_part', 'argument_description': 'Filters for work belonging to any of the provided parts', 'argument_type': 'array of strings', 'examples': ['FEAT-123', 'ENH-123', 'PROD-123', 'CAPL-123']}, {'argument_name': 'created_by', 'argument_description': 'Filters for work created by any of these users', 'argument_type': 'array of strings', 'examples': ['DEVU-123']}, {'argument_name': 'issue.priority', 'argument_description': 'Filters for issues with any of the provided priorities. Allowed values: p0, p1, p2,p3', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'issue.rev_orgs', 'argument_description': 'Filters for issues with any of the provided Rev organizations', 'argument_type': 'array of strings', 'examples': ['REV-123']}, {'argument_name': 'limit', 'argument_description': "The maximum number of works to return. The default is '50'", 'argument_type': 'integer (int32)', 'examples': []}, {'argument_name': 'owned_by', 'argument_description': 'Filters for work owned by any of these users', 'argument_type': 'array of strings', 'examples': ['DEVU-123']}, {'argument_name': 'stage.name', 'argument_description': 'Filters for records in the provided stage(s) by name', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'ticket.needs_response', 'argument_description': 'Filters for tickets that need a response', 'argument_type': 'boolean', 'examples': []}, {'argument_name': 'ticket.rev_org', 'argument_description': 'Filters for tickets associated with any of the provided Rev organizations', 'argument_type': 'array of strings', 'examples': ['REV-123']}, {'argument_name': 'ticket.severity', 'argument_description': 'Filters for tickets with any of the provided severities. Allowed values:blocker,high, low, medium', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'ticket.source_channel', 'argument_description': 'Filters for tickets with any of the provided source channels', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'ticket.source_channel', 'argument_description': 'Filters for tickets with any of the provided source channels', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'type', 'argument_description': 'Filters for work of the provided types. Allowed values issue, ticket, task', 'argument_type': 'array of strings', 'examples': []}]}, {'tool': 'add_work_items_to_sprint', 'description': 'Adds the given work items to the sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_description': 'A list of work item IDs to be added to the sprint.', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'sprint_id', 'argument_description': 'The ID of the sprint to which the work items should be added', 'argument_type': 'str', 'examples': []}]}, {'tool': 'summarize_objects', 'description': 'Summarizes a list of objects. The logic of how to summarize a particular object type is an internal implementation detail.', 'arguments': [{'argument_name': 'objects', 'argument_description': 'List of objects to summarize', 'argument_type': 'array of objects', 'examples': []}]}] Answer: [/INST] { "tools": [ { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-456" } }, { "tool": "summarize_objects", "arguments": { "objects": "$PREV[0]" } }, { "tool": "works_list", "arguments": { "applies_to_part": "$PREV[1]", "created_by": [], "issue.priority": [], "issue.rev_orgs": [], "limit": 50, "owned_by": [], "stage.name": [], "ticket.needs_response": false, "ticket.rev_org": [], "ticket.severity": [], "ticket.source_channel": [], "type": [] } }, { "tool": "add_work_items_to_sprint", "arguments": { "work_ids": "$PREV[2]", "sprint_id": "" } } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [[{'query': 'Get details for part ID PROD-789, update its information, and delete the part with ID PROD-xyz123.', 'solutions': [{'tool': 'parts.get', 'arguments': [{'argument_name': 'id', 'argument_value': 'PROD-789'}]}, {'tool': 'parts.update', 'arguments': []}, {'tool': 'parts.delete', 'arguments': [{'argument_name': 'id', 'argument_value': 'PROD-xyz123'}]}]}], [{'query': 'Create a new part, retrieve its information, update its details, delete a different part, and list parts created by DEVU-78901.', 'solutions': [{'tool': 'parts.create', 'arguments': []}, {'tool': 'parts.get', 'arguments': [{'argument_name': 'id', 'argument_value': '$$PREV[0]'}]}, {'tool': 'parts.update', 'arguments': []}, {'tool': 'parts.delete', 'arguments': [{'argument_name': 'id', 'argument_value': '$$PREV[2]'}]}, {'tool': 'parts.list', 'arguments': [{'argument_name': 'created_by', 'argument_value': ['DEVU-78901']}]}]}], [{'query': 'Create a new part, update its details, and fetch the list of parts owned by DEVU-45678.', 'solutions': [{'tool': 'parts.create', 'arguments': []}, {'tool': 'parts.update', 'arguments': []}, {'tool': 'parts.list', 'arguments': [{'argument_name': 'owned_by', 'argument_value': ['DEVU-45678']}]}]}]] Current Query : Get details for part ID PROD-789, update its information, and delete the part with ID PROD-xyz123. Current Solution : { "tools": [ { "tool": "parts.get", "arguments": [ { "argument_name": "id", "argument_value": "PROD-789" } ] }, { "tool": "parts.update", "arguments": [ { "argument_name": "id", "argument_value": "PROD-789" } ] }, { "tool": "parts.delete", "arguments": [ { "argument_name": "id", "argument_value": "PROD-xyz123" } ] } ] } Answer: [/INST] { "tools": [ { "tool": "parts.get", "arguments": [ { "argument_name": "id", "argument_value": "PROD-789" } ] }, { "tool": "parts.update", "arguments": [] }, { "tool": "parts.delete", "arguments": [ { "argument_name": "id", "argument_value": "PROD-xyz123" } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [[{'query': 'Create a new timeline entry, retrieve the list of entries for ObjectID_Gamma_456, and update an entry.', 'solutions': [{'tool_name': 'timeline-entries.create', 'arguments': []}, {'tool_name': 'timeline-entries.list', 'arguments': [{'argument_name': 'object', 'argument_value': 'ObjectID_Gamma_456'}]}, {'tool_name': 'timeline-entries.update', 'arguments': []}]}], [{'query': 'Create a new timeline entry, retrieve the entry data for EntryID_Beta_789, and then delete the entry.', 'solutions': [{'tool_name': 'timeline-entries.create', 'arguments': []}, {'tool_name': 'timeline-entries.get', 'arguments': [{'argument_name': 'id', 'argument_value': 'EntryID_Beta_789'}]}, {'tool_name': 'timeline-entries.delete', 'arguments': []}]}], [{'query': 'Create a new task, fetch details for TaskID_AlphaNum_789, and mark the task as completed.', 'solutions': [{'tool_name': 'tasks.create', 'arguments': []}, {'tool_name': 'tasks.getDetails', 'arguments': [{'argument_name': 'task_id', 'argument_value': 'TaskID_AlphaNum_789'}]}, {'tool_name': 'tasks.markCompleted', 'arguments': []}]}]] Current Query : Create a new timeline entry, retrieve the list of entries for ObjectID_Gamma_456, and update an entry. Current Solution : { "tools": [ { "tool": "timeline-entries.create", "arguments": [] }, { "tool": "timeline-entries.list", "arguments": [ { "argument_name": "object", "argument_description": "The ID of the object to list timeline entries for.", "argument_type": "string", "examples": "PROD-12345" }, { "argument_name": "cursor", "argument_description": "The cursor to resume iteration from. If not provided, then iteration\nstarts from the beginning.\n", "argument_type": "string", "examples": [] }, { "argument_name": "limit", "argument_description": "The maximum number of entries to return. If not set, then this\ndefaults to `50`.\n", "argument_type": "integer", "examples": [] }, { "argument_name": "mode", "argument_description": "The iteration mode to use, otherwise if not set, then \"after\" is\nused.\n", "argument_type": "", "examples": [] }, { "argument_name": "visibility", "argument_description": "The visibility of the timeline entries to filter for. Note this is a\nstrict filter, such that only entries with the exact visibilities\nspecified will be returned.\n", "argument_type": "array", "examples": [] } ] }, { "tool": "timeline-entries.update", "arguments": [] } ] } Answer: [/INST] { "tools": [ { "tool": "timeline-entries.create", "arguments": [] }, { "tool": "timeline-entries.list", "arguments": [ { "argument_name": "object", "argument_value": "ObjectID_Gamma_456" } ] }, { "tool": "timeline-entries.update", "arguments": [] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'Query': 'Summarize issues similar to don:core:dvrv-us-1:devo/0:issue/1', 'Solution': [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'don:core:dvrv-us-1:devo/0:issue/1'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[0]'}]}]}, {'Query': 'Get all work items similar to TKT-123, summarize them, create issues from that summary, and prioritize them', 'Solution': [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'TKT-123'}]}, {'tool_name': 'create_actionable_tasks_from_text', 'arguments': [{'argument_name': 'text', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'What are my all issues in the triage stage under part FEAT-123? Summarize them.', 'Solution': [{'tool_name': 'who_am_i', 'arguments': []}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'stage.name', 'argument_value': 'triage'}, {'argument_name': 'applies_to_part', 'argument_value': 'FEAT-123'}, {'argument_name': 'owned_by', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}] Current Query : Summarize ideas that are similar to ABCD1234:devo/2:issue/32 Current Solution : [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'ABCD1234:devo/2:issue/32'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$PREV[0]'}]}] Answer: [/INST] [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'ABCD1234:devo/2:issue/32'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[0]'}]}] </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [[{'query': 'Retrieve information for REV-65432 organization, update its details, and get the list of Rev organizations created after 2023-03-15.', 'solutions': [{'tool': 'rev-orgs.get', 'arguments': [{'argument_name': 'id', 'argument_value': 'REV-65432'}]}, {'tool': 'rev-orgs.update', 'arguments': []}, {'tool': 'rev-orgs.list', 'arguments': [{'argument_name': 'created_date.after', 'argument_value': '2023-03-15T00:00:00.000Z'}]}]}], [{'query': 'Retrieve the information for REV-12345 organization, update its details, and get the list of Rev organizations created after 2023-01-01.', 'solutions': [{'tool': 'rev-orgs.get', 'arguments': [{'argument_name': 'id', 'argument_value': 'REV-12345'}]}, {'tool': 'rev-orgs.update', 'arguments': []}, {'tool': 'rev-orgs.list', 'arguments': [{'argument_name': 'created_date.after', 'argument_value': '2023-01-01T00:00:00.000Z'}]}]}], [{'query': 'Get the list of Rev organizations created before 2023-06-30, retrieve details for REV-98765 organization, and delete the organization with ID REV-xyz987.', 'solutions': [{'tool': 'rev-orgs.list', 'arguments': [{'argument_name': 'created_date.before', 'argument_value': '2023-06-30T23:59:59.999Z'}]}, {'tool': 'rev-orgs.get', 'arguments': [{'argument_name': 'id', 'argument_value': 'REV-98765'}]}, {'tool': 'rev-orgs.delete', 'arguments': [{'argument_name': 'id', 'argument_value': 'REV-xyz987'}]}]}]] Current Query : Retrieve information for REV-65432 organization, update its details, and get the list of Rev organizations created after 2023-03-15. Current Solution : [ { "tool": "rev-orgs.get", "arguments": [ { "argument_name": "account", "argument_value": "$PREV[0]" }, { "argument_name": "id", "argument_value": "REV-65432" } ] }, { "tool": "rev-orgs.update", "arguments": [ { "argument_name": "id", "argument_value": "REV-65432" } ] }, { "tool": "rev-orgs.list", "arguments": [ { "argument_name": "created_date.after", "argument_value": "2023-03-15" } ] } ] Answer: [/INST] { "tool": "rev-orgs.get", "arguments": [ { "argument_name": "id", "argument_value": "REV-65432" } ] }, { "tool": "rev-orgs.update", "arguments": [ { "argument_name": "id", "argument_value": "REV-65432" } ] }, { "tool": "rev-orgs.list", "arguments": [ { "argument_name": "created_date.after", "argument_value": "2023-03-15T00:00:00.000Z" } ] } </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Get all work items similar to TKT-456, TKT-458, T TKT-460, summarize them, create issues from that summary, and prioritize them APIs: [{'tool': 'get_similar_work_items', 'description': 'Returns a list of work items that are similar to the given work item', 'arguments': [{'argument_name': 'work_id', 'argument_description': 'The ID of the work item for which you want to find similar items', 'argument_type': 'string', 'examples': []}]}, {'tool': 'works_list', 'description': 'Returns a list of work items matching the request', 'arguments': [{'argument_name': ' applies_to_part', 'argument_description': 'Filters for work belonging to any of the provided parts', 'argument_type': 'array of strings', 'examples': ['FEAT-123', 'ENH-123', 'PROD-123', 'CAPL-123']}, {'argument_name': 'created_by', 'argument_description': 'Filters for work created by any of these users', 'argument_type': 'array of strings', 'examples': ['DEVU-123']}, {'argument_name': 'issue.priority', 'argument_description': 'Filters for issues with any of the provided priorities. Allowed values: p0, p1, p2,p3', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'issue.rev_orgs', 'argument_description': 'Filters for issues with any of the provided Rev organizations', 'argument_type': 'array of strings', 'examples': ['REV-123']}, {'argument_name': 'limit', 'argument_description': "The maximum number of works to return. The default is '50'", 'argument_type': 'integer (int32)', 'examples': []}, {'argument_name': 'owned_by', 'argument_description': 'Filters for work owned by any of these users', 'argument_type': 'array of strings', 'examples': ['DEVU-123']}, {'argument_name': 'stage.name', 'argument_description': 'Filters for records in the provided stage(s) by name', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'ticket.needs_response', 'argument_description': 'Filters for tickets that need a response', 'argument_type': 'boolean', 'examples': []}, {'argument_name': 'ticket.rev_org', 'argument_description': 'Filters for tickets associated with any of the provided Rev organizations', 'argument_type': 'array of strings', 'examples': ['REV-123']}, {'argument_name': 'ticket.severity', 'argument_description': 'Filters for tickets with any of the provided severities. Allowed values:blocker,high, low, medium', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'ticket.source_channel', 'argument_description': 'Filters for tickets with any of the provided source channels', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'ticket.source_channel', 'argument_description': 'Filters for tickets with any of the provided source channels', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'type', 'argument_description': 'Filters for work of the provided types. Allowed values issue, ticket, task', 'argument_type': 'array of strings', 'examples': []}]}, {'tool': 'add_work_items_to_sprint', 'description': 'Adds the given work items to the sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_description': 'A list of work item IDs to be added to the sprint.', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'sprint_id', 'argument_description': 'The ID of the sprint to which the work items should be added', 'argument_type': 'str', 'examples': []}]}, {'tool': 'summarize_objects', 'description': 'Summarizes a list of objects. The logic of how to summarize a particular object type is an internal implementation detail.', 'arguments': [{'argument_name': 'objects', 'argument_description': 'List of objects to summarize', 'argument_type': 'array of objects', 'examples': []}]}] Answer: [/INST] { "tools": [ { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-456" } }, { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-458" } }, { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-460" } }, { "tool": "summarize_objects", "arguments": { "objects": [ "$PREV[0]", "$PREV[1]", "$PREV[2]" ] } }, { "tool": "works_list", "arguments": { "issue.priority": ["p0", "p1", "p2", "p3"] } }, { "tool": "add_work_items_to_sprint", "arguments": { "work_ids": ["$PREV[3]"], "sprint_id": "sprint_id" } } ] } </s>

<s>[INST] Given a query, solution and available APIs, optimize the number of APIs calls in the solution. Do not use API calls unless absolutely necessary. Look for redundancy and check for mistakes in the combination of API calls. Return current solution if no changes are necessary. Current Query : Perform a comprehensive update on custom field options with id 789 and get details for a set of issues within a Forge app. Current Solution : { "APIs": [ { "tool": "issue_custom_field_options", "arguments": [ { "argument_name": "customFieldOption/:id", "argument_value": "789" } ] }, { "tool": "issue_custom_field_options_(apps)", "arguments": [ { "argument_name": "field/:fieldKey/option", "argument_value": "<option_details>" } ] }, { "tool": "issue_custom_field_values_(apps)", "arguments": [ { "argument_name": "app/field/value", "argument_value": "Forge" } ] } ] } Available APIs : [{'tool': 'issue_custom_field_values_(apps)', 'description': 'This resource represents the values of custom fields added by [Forge apps](https://developer.atlassian.com/platform/forge/). Use it to update the value of a custom field on issues.', 'arguments': [{'argument_name': 'app/field/value', 'argument description': 'Updates the value of one or more custom fields on one or more issues. Combinations of custom field and issue should be unique within the request. Custom fields can only be updated by the Forge app that created them.\n\n**[Permissions](#permissions) required:** Only the app that created the custom field can update its values with this operation.', 'argument type': 'application/json'}, {'argument_name': 'app/field/:fieldIdOrKey/value', 'argument description': 'Updates the value of a custom field on one or more issues. Custom fields can only be updated by the Forge app that created them.\n\n**[Permissions](#permissions) required:** Only the app that created the custom field can update its values with this operation.', 'argument type': 'application/json'}]}, {'tool': 'ui_modifications_(apps)', 'description': 'UI modifications is a feature available for **Forge apps only**. It enables Forge apps to control how selected Jira fields behave on the following views: global issue create, issue view. For example: hide specific fields, set them as required, etc.', 'arguments': [{'argument_name': 'uiModifications', 'argument description': 'Gets UI modifications. UI modifications can only be retrieved by Forge apps.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'uiModifications', 'argument description': 'Creates a UI modification. UI modification can only be created by Forge apps.\n\nEach app can define up to 3000 UI modifications. Each UI modification can define up to 1000 contexts. The same context can be assigned to maximum 100 UI modifications.\n\n**[Permissions](#permissions) required:**\n\n * *None* if the UI modification is created without contexts.\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for one or more projects, if the UI modification is created with contexts.', 'argument type': 'application/json'}, {'argument_name': 'uiModifications/:uiModificationId', 'argument description': 'Updates a UI modification. UI modification can only be updated by Forge apps.\n\nEach UI modification can define up to 1000 contexts. The same context can be assigned to maximum 100 UI modifications.\n\n**[Permissions](#permissions) required:**\n\n * *None* if the UI modification is created without contexts.\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for one or more projects, if the UI modification is created with contexts.', 'argument type': 'application/json'}, {'argument_name': 'uiModifications/:uiModificationId', 'argument description': 'Deletes a UI modification. All the contexts that belong to the UI modification are deleted too. UI modification can only be deleted by Forge apps.\n\n**[Permissions](#permissions) required:** None.', 'argument type': ''}]}, {'tool': 'issue_custom_field_configuration_(apps)', 'description': 'This resource represents configurations stored against a custom field context by a [Forge app](https://developer.atlassian.com/platform/forge/). Configurations are information used by the Forge app at runtime to determine how to handle or process the data in a custom field in a given context. Use this resource to set and read configurations.', 'arguments': [{'argument_name': 'app/field/:fieldIdOrKey/context/configuration', 'argument description': 'Returns a [paginated](#pagination) list of configurations for a custom field created by a [Forge app](https://developer.atlassian.com/platform/forge/).\n\nThe result can be filtered by one of these criteria:\n\n * `id`.\n * `fieldContextId`.\n * `issueId`.\n * `projectKeyOrId` and `issueTypeId`.\n\nOtherwise, all configurations are returned.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg). Jira permissions are not required for the Forge app that created the custom field.', 'argument type': 'application/json'}, {'argument_name': 'app/field/:fieldIdOrKey/context/configuration', 'argument description': 'Update the configuration for contexts of a custom field created by a [Forge app](https://developer.atlassian.com/platform/forge/).\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg). Jira permissions are not required for the Forge app that created the custom field.', 'argument type': 'application/json'}]}, {'tool': 'issue_fields', 'description': 'This resource represents issue fields, both system and custom fields. Use it to get fields, field configurations, and create custom fields.', 'arguments': [{'argument_name': 'field', 'argument description': 'Returns system and custom issue fields according to the following rules:\n\n * Fields that cannot be added to the issue navigator are always returned.\n * Fields that cannot be placed on an issue screen are always returned.\n * Fields that depend on global Jira settings are only returned if the setting is enabled. That is, timetracking fields, subtasks, votes, and watches.\n * For all other fields, this operation only returns the fields that the user has permission to view (that is, the field is used in at least one project that the user has *Browse Projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for.)\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'field', 'argument description': 'Creates a custom field.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/search', 'argument description': 'Returns a [paginated](#pagination) list of fields for Classic Jira projects. The list can include:\n\n * all fields\n * specific fields, by defining `id`\n * fields that contain a string in the field name or description, by defining `query`\n * specific fields that contain a string in the field name or description, by defining `id` and `query`\n\nOnly custom fields can be queried, `type` must be set to `custom`.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/search/trashed', 'argument description': 'Returns a [paginated](#pagination) list of fields in the trash. The list may be restricted to fields whose field name or description partially match a string.\n\nOnly custom fields can be queried, `type` must be set to `custom`.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:fieldId', 'argument description': 'Updates a custom field.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:fieldId/contexts', 'argument description': 'Returns a [paginated](#pagination) list of the contexts a field is used in. Deprecated, use [ Get custom field contexts](#api-rest-api-3-field-fieldId-context-get).\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:id', 'argument description': 'Deletes a custom field. The custom field is deleted whether it is in the trash or not. See [Edit or delete a custom field](https://confluence.atlassian.com/x/Z44fOw) for more information on trashing and deleting custom fields.\n\nThis operation is [asynchronous](#async). Follow the `location` link in the response to determine the status of the task and use [Get task](#api-rest-api-3-task-taskId-get) to obtain subsequent updates.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': ''}, {'argument_name': 'field/:id/restore', 'argument description': 'Restores a custom field from trash. See [Edit or delete a custom field](https://confluence.atlassian.com/x/Z44fOw) for more information on trashing and deleting custom fields.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:id/trash', 'argument description': 'Moves a custom field to trash. See [Edit or delete a custom field](https://confluence.atlassian.com/x/Z44fOw) for more information on trashing and deleting custom fields.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}]}] Final solution (same format as current solution): [/INST] { "APIs": [ { "tool": "issue_custom_field_options", "arguments": [ { "argument_name": "customFieldOption/:id", "argument_value": "789" } ] }, { "tool": "issue_custom_field_options_(apps)", "arguments": [ { "argument_name": "field/:fieldKey/option", "argument_value": "<option_details>" } ] }, { "tool": "issue_custom_field_values_(apps)", "arguments": [ { "argument_name": "app/field/value", "argument_value": "Forge" } ] } ] } </s>

<s>[INST] Given a query, solution and available APIs, optimize the number of APIs calls in the solution. Do not use API calls unless absolutely necessary. Look for redundancy and check for mistakes in the combination of API calls. Return current solution if no changes are necessary. Current Query : Set Pheonix avatar for a project ABC123, return project category, and update the details of the project category and return the count of issues assigned to component XYZ Current Solution : [] Available APIs : [{'tool': 'project_avatars', 'description': 'This resource represents avatars associated with a project. Use it to get, load, set, and remove project avatars.', 'arguments': [{'argument_name': 'project/:projectIdOrKey/avatar', 'argument description': 'Sets the avatar displayed for a project.\n\nUse [Load project avatar](#api-rest-api-3-project-projectIdOrKey-avatar2-post) to store avatars against the project, before using this operation to set the displayed avatar.\n\n**[Permissions](#permissions) required:** *Administer projects* [project permission](https://confluence.atlassian.com/x/yodKLg).', 'argument type': 'application/json'}, {'argument_name': 'project/:projectIdOrKey/avatar/:id', 'argument description': 'Deletes a custom avatar from a project. Note that system avatars cannot be deleted.\n\n**[Permissions](#permissions) required:** *Administer projects* [project permission](https://confluence.atlassian.com/x/yodKLg).', 'argument type': ''}, {'argument_name': 'project/:projectIdOrKey/avatar2', 'argument description': 'Loads an avatar for a project.\n\nSpecify the avatar\'s local file location in the body of the request. Also, include the following headers:\n\n * `X-Atlassian-Token: no-check` To prevent XSRF protection blocking the request, for more information see [Special Headers](#special-request-headers).\n * `Content-Type: image/image type` Valid image types are JPEG, GIF, or PNG.\n\nFor example: \n`curl --request POST `\n\n`--user [email protected]: `\n\n`--header \'X-Atlassian-Token: no-check\' `\n\n`--header \'Content-Type: image/\' `\n\n`--data-binary "" `\n\n`--url \'https://your-domain.atlassian.net/rest/api/3/project/{projectIdOrKey}/avatar2\'`\n\nThe avatar is cropped to a square. If no crop parameters are specified, the square originates at the top left of the image. The length of the square\'s sides is set to the smaller of the height or width of the image.\n\nThe cropped image is then used to create avatars of 16x16, 24x24, 32x32, and 48x48 in size.\n\nAfter creating the avatar use [Set project avatar](#api-rest-api-3-project-projectIdOrKey-avatar-put) to set it as the project\'s displayed avatar.\n\n**[Permissions](#permissions) required:** *Administer projects* [project permission](https://confluence.atlassian.com/x/yodKLg).', 'argument type': '*/*'}, {'argument_name': 'project/:projectIdOrKey/avatars', 'argument description': 'Returns all project avatars, grouped by system and custom avatars.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project.', 'argument type': 'application/json'}]}, {'tool': 'avatars', 'description': 'This resource represents system and custom avatars. Use it to obtain the details of system or custom avatars, add and remove avatars from a project or issue type, and obtain avatar images.', 'arguments': [{'argument_name': 'avatar/:type/system', 'argument description': 'Returns a list of system avatar details by owner type, where the owner types are issue type, project, or user.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'universal_avatar/type/:type/owner/:entityId', 'argument description': 'Returns the system and custom avatars for a project or issue type.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * for custom project avatars, *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project the avatar belongs to.\n * for custom issue type avatars, *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for at least one project the issue type is used in.\n * for system avatars, none.', 'argument type': 'application/json'}, {'argument_name': 'universal_avatar/type/:type/owner/:entityId', 'argument description': 'Loads a custom avatar for a project or issue type.\n\nSpecify the avatar\'s local file location in the body of the request. Also, include the following headers:\n\n * `X-Atlassian-Token: no-check` To prevent XSRF protection blocking the request, for more information see [Special Headers](#special-request-headers).\n * `Content-Type: image/image type` Valid image types are JPEG, GIF, or PNG.\n\nFor example: \n`curl --request POST `\n\n`--user [email protected]: `\n\n`--header \'X-Atlassian-Token: no-check\' `\n\n`--header \'Content-Type: image/\' `\n\n`--data-binary "" `\n\n`--url \'https://your-domain.atlassian.net/rest/api/3/universal_avatar/type/{type}/owner/{entityId}\'`\n\nThe avatar is cropped to a square. If no crop parameters are specified, the square originates at the top left of the image. The length of the square\'s sides is set to the smaller of the height or width of the image.\n\nThe cropped image is then used to create avatars of 16x16, 24x24, 32x32, and 48x48 in size.\n\nAfter creating the avatar use:\n\n * [Update issue type](#api-rest-api-3-issuetype-id-put) to set it as the issue type\'s displayed avatar.\n * [Set project avatar](#api-rest-api-3-project-projectIdOrKey-avatar-put) to set it as the project\'s displayed avatar.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': '*/*'}, {'argument_name': 'universal_avatar/type/:type/owner/:owningObjectId/avatar/:id', 'argument description': 'Deletes an avatar from a project or issue type.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': ''}, {'argument_name': 'universal_avatar/view/type/:type', 'argument description': 'Returns the default project or issue type avatar image.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'universal_avatar/view/type/:type/avatar/:id', 'argument description': 'Returns a project or issue type avatar image by ID.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * For system avatars, none.\n * For custom project avatars, *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project the avatar belongs to.\n * For custom issue type avatars, *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for at least one project the issue type is used in.', 'argument type': 'application/json'}, {'argument_name': 'universal_avatar/view/type/:type/owner/:entityId', 'argument description': 'Returns the avatar image for a project or issue type.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * For system avatars, none.\n * For custom project avatars, *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project the avatar belongs to.\n * For custom issue type avatars, *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for at least one project the issue type is used in.', 'argument type': 'application/json'}]}, {'tool': 'project_components', 'description': 'This resource represents project components. Use it to get, create, update, and delete project components. Also get components for project and get a count of issues by component.', 'arguments': [{'argument_name': 'component', 'argument description': 'Creates a component. Use components to provide containers for issues within a project. Use components to provide containers for issues within a project.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** *Administer projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project in which the component is created or *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'component/:id', 'argument description': 'Returns a component.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for project containing the component.', 'argument type': 'application/json'}, {'argument_name': 'component/:id', 'argument description': 'Updates a component. Any fields included in the request are overwritten. If `leadAccountId` is an empty string ("") the component lead is removed.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** *Administer projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project containing the component or *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'component/:id', 'argument description': 'Deletes a component.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** *Administer projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project containing the component or *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': ''}, {'argument_name': 'component/:id/relatedIssueCounts', 'argument description': 'Returns the counts of issues assigned to the component.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'project/:projectIdOrKey/component', 'argument description': 'Returns a [paginated](#pagination) list of all components in a project. See the [Get project components](#api-rest-api-3-project-projectIdOrKey-components-get) resource if you want to get a full list of versions without pagination.\n\nIf your project uses Compass components, this API will return a list of Compass components that are linked to issues in that project.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** *Browse Projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project.', 'argument type': 'application/json'}, {'argument_name': 'project/:projectIdOrKey/components', 'argument description': 'Returns all components in a project. See the [Get project components paginated](#api-rest-api-3-project-projectIdOrKey-component-get) resource if you want to get a full list of components with pagination.\n\nIf your project uses Compass components, this API will return a paginated list of Compass components that are linked to issues in that project.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** *Browse Projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project.', 'argument type': 'application/json'}]}, {'tool': 'project_categories', 'description': 'This resource represents project categories. Use it to create, update, and delete project categories as well as obtain a list of all project categories and details of individual categories. For more information on managing project categories, see [Adding, assigning, and deleting project categories](https://confluence.atlassian.com/x/-A5WMg).', 'arguments': [{'argument_name': 'projectCategory', 'argument description': 'Returns all project categories.\n\n**[Permissions](#permissions) required:** Permission to access Jira.', 'argument type': 'application/json'}, {'argument_name': 'projectCategory', 'argument description': 'Creates a project category.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'projectCategory/:id', 'argument description': 'Returns a project category.\n\n**[Permissions](#permissions) required:** Permission to access Jira.', 'argument type': 'application/json'}, {'argument_name': 'projectCategory/:id', 'argument description': 'Updates a project category.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'projectCategory/:id', 'argument description': 'Deletes a project category.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': ''}]}] Final solution (same format as current solution): [/INST] "query": "Set Pheonix avatar for a project ABC123, return project category, and update the details of the project category and return the count of issues assigned to component XYZ", "solution": [ { "tool_name": "project_avatars", "arguments": [ { "argument_name": "project/:projectIdOrKey/avatar", "argument_value": "Pheonix" } ] }, { "tool_name": "project_categories", "arguments": [ </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Retrieve the current announcement banner configuration including required permissions for access, and then get details of all application roles in Jira. APIs: [{'tool': 'announcement_banner', 'description': 'This resource represents an announcement banner. Use it to retrieve and update banner configuration.', 'arguments': [{'argument_name': 'announcementBanner', 'argument description': 'Returns the current announcement banner configuration.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'announcementBanner', 'argument description': 'Updates the announcement banner configuration.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}]}, {'tool': 'server_info', 'description': 'This resource provides information about the Jira instance.', 'arguments': [{'argument_name': 'serverInfo', 'argument description': 'Returns information about the Jira instance.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}]}, {'tool': 'application_roles', 'description': 'This resource represents application roles. Use it to get details of an application role or all application roles.', 'arguments': [{'argument_name': 'applicationrole', 'argument description': 'Returns all application roles. In Jira, application roles are managed using the [Application access configuration](https://confluence.atlassian.com/x/3YxjL) page.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'applicationrole/:key', 'argument description': 'Returns an application role.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}]}, {'tool': 'license_metrics', 'description': 'This resource represents license metrics. Use it to get available metrics for Jira licences.', 'arguments': [{'argument_name': 'instance/license', 'argument description': 'Returns licensing information about the Jira instance.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'license/approximateLicenseCount', 'argument description': 'Returns the approximate number of user accounts across all Jira licenses. Note that this information is cached with a 7-day lifecycle and could be stale at the time of call.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'license/approximateLicenseCount/product/:applicationKey', 'argument description': 'Returns the total approximate number of user accounts for a single Jira license. Note that this information is cached with a 7-day lifecycle and could be stale at the time of call.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}]}] Answer: [/INST] { "APIs": [ { "tool": "announcement_banner", "arguments": [ { "argument_name": "announcementBanner", "argument_value": "$PREV[0]" } ] }, { "tool": "application_roles", "arguments": [ { "argument_name": "applicationrole", "argument_value": "" } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [[{'query': "Retrieve all tokens issued for the application example_app and update the metadata for a token issued under the 'Dev' organization.", 'solutions': [{'tool': 'auth-tokens.list', 'arguments': [{'argument_name': 'client_id', 'argument_value': 'example_app'}]}, {'tool': 'auth-tokens.update', 'arguments': []}]}, {'query': 'Revoke all tokens of a specific type issued by the authenticated user.', 'solutions': [{'tool': 'auth-tokens.self.delete', 'arguments': []}]}], [{'query': "Retrieve tokens issued for the application 'app_456' and update token metadata for a specific token issued under a Dev organization.", 'solutions': [{'tool': 'auth-tokens.list', 'arguments': [{'argument_name': 'client_id', 'argument_value': 'app_456'}]}, {'tool': 'auth-tokens.update', 'arguments': []}]}], [{'query': "Retrieve all tokens for the subject 'user123' of a specific type and update the metadata of a token issued under a Dev organization.", 'solutions': [{'tool': 'auth-tokens.list', 'arguments': [{'argument_name': 'subject', 'argument_value': 'user123'}]}, {'tool': 'auth-tokens.update', 'arguments': []}]}]] Current Query : Retrieve all tokens issued for the application example_app and update the metadata for a token issued under the 'Dev' organization. Current Solution : { "tools": [ { "tool": "auth-tokens.list", "arguments": [ { "argument_name": "client_id", "argument_value": "example_app" } ] }, { "tool": "auth-tokens.update", "arguments": [] } ] } Answer: [/INST] { "tools": [ { "tool": "auth-tokens.list", "arguments": [ { "argument_name": "client_id", "argument_value": "example_app" } ] }, { "tool": "auth-tokens.update", "arguments": [] } ] } </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Find all high severity tickets coming in from messenger from customer AbraKaDabra and generate a summary of them. Add the summary to my current sprint. APIs: [{ "tool": "who_am_i", "description": "Returns the ID of the current user", "arguments": [ { "argument_name": "", "argument_description": "", "argument_type": "", "examples": [] } ] },{'tool': 'get_sprint_id', 'description': 'Returns the ID of the current sprint', 'arguments': [{'argument_name': '', 'argument_description': '', 'argument_type': '', 'examples': []}]}, {'tool': 'add_work_items_to_sprint', 'description': 'Adds the given work items to the sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_description': 'A list of work item IDs to be added to the sprint.', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'sprint_id', 'argument_description': 'The ID of the sprint to which the work items should be added', 'argument_type': 'str', 'examples': []}]}, {'tool': 'summarize_objects', 'description': 'Summarizes a list of objects. The logic of how to summarize a particular object type is an internal implementation detail.', 'arguments': [{'argument_name': 'objects', 'argument_description': 'List of objects to summarize', 'argument_type': 'array of objects', 'examples': []}]}, {'tool': 'prioritize_objects', 'description': 'Summarizes a list of objects. The logic of how to summarize a particular object type is an internal implementation detail.', 'arguments': [{'argument_name': 'objects', 'argument_description': 'List of objects to summarize', 'argument_type': 'array of objects', 'examples': []}]}] Answer: [/INST] { "APIs": [ { "tool": "get_sprint_id", "arguments": [] }, { "tool": "summarize_objects", "arguments": [ { "argument_name": "objects", "argument_description": "List of objects to summarize", "argument_type": "array of objects", "examples": [] } ] }, { "tool": "prioritize_objects", "arguments": [ { "argument_name": "objects", "argument_description": "List of objects to summarize", "argument_type": "array of objects", "examples": [] } ] }, { "tool": "add_work_items_to_sprint", "arguments": [ { "argument_name": "work_ids", "argument_description": "A list of work item IDs to be added to the sprint.", "argument_type": "array of strings", "examples": [] }, { "argument_name": "sprint_id", "argument_description": "The ID of the sprint to which the work items should be added", "argument_type": "str", "examples": [] } ] } ] } </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Create an enterprise authentication connection, retrieve its details, and then delete the connection. APIs: [{'tool': 'dev-orgs.auth-connections.delete', 'description': 'Deletes an authentication connection. Only enterprise connections which\nare explicitly set up for a Dev organization can be deleted. Default\nconnections can not be deleted using this method.\n', 'arguments': []}, {'tool': 'dev-orgs.auth-connections.get', 'description': 'Retrieves the details for an authentication connection.', 'arguments': [{'argument_name': 'id', 'argument_description': 'ID of the authentication connection.', 'argument_type': 'string', 'examples': []}]}, {'tool': 'dev-orgs.auth-connections.update', 'description': 'Updates an authentication connection.', 'arguments': []}, {'tool': 'dev-orgs.auth-connections.create', 'description': 'Creates a new enterprise authentication connection for a Dev\norganization. This authentication connection will not be enabled by\ndefault for the organization and the user will need to explicitly\nenable this. Keep in mind that at a time, only one authentication\nconnection can be enabled for a Dev organization. At present, only 5\nenterprise connections can be created by an organization.\n', 'arguments': []}] Answer: [/INST] { "tools": [ { "tool_name": "dev-orgs.auth-connections.create", "arguments": [] }, { "tool_name": "dev-orgs.auth-connections.get", "arguments": [ { "argument_name": "id", "argument_value": "$$PREV[0]" } ] }, { "tool_name": "dev-orgs.auth-connections.delete", "arguments": [] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'query': 'Retrieve default columns for the issue navigator, manage notification schemes for scheme id LM10(get existing, create new, update, and customize), and update default columns if necessary.', 'solution': [{'tool_name': 'issue_navigator_settings', 'arguments': [{'argument_name': 'get_default_columns', 'argument_value': ''}]}, {'tool_name': 'issue_notification_schemes', 'arguments': [{'argument_name': 'create_notification_scheme', 'argument_value': {'scheme_details': 'Provide details for the new notification scheme.'}}, {'argument_name': 'update_notification_scheme', 'argument_value': {'scheme_id': 'LM10'}}, {'argument_name': 'add_notifications_to_scheme', 'argument_value': {'scheme_id': 'LM10'}}]}]}, {'query': 'Perform a comprehensive update on custom field options with id 789 and get details for a set of issues within a Forge app.', 'solutions': [{'tool': 'issue_custom_field_options', 'arguments': [{'argument_name': 'customFieldOption/:id', 'argument_value': '789'}]}, {'tool': 'issue_custom_field_options_(apps)', 'arguments': [{'argument_name': 'field/:fieldKey/option', 'argument_value': '<option_details>'}]}, {'tool': 'issue_custom_field_values_(apps)', 'arguments': [{'argument_name': 'app/field/value', 'argument_value': 'Forge'}]}]}, {'query': 'Retrieve details of an issue security level and create a new issue security scheme with security scheme levels and members.', 'solutions': [{'tool_name': 'issue_security_level', 'arguments': [{'argument_name': 'securitylevel/:id', 'argument_value': {'argument_type': 'application/json'}}]}, {'tool_name': 'issue_security_schemes', 'arguments': [{'argument_name': 'issuesecurityschemes', 'argument_value': {'argument_type': 'application/json'}}]}]}] Current Query : Retrieve default columns for the issue navigator, manage notification schemes for scheme id LM10(get existing, create new, update, and customize), and update default columns if necessary. Current Solution : { "tools": [ { "tool": "issue_navigator_settings", "arguments": [ { "argument_name": "get_default_columns", "argument_value": "" } ] }, { "tool": "issue_notification_schemes", "arguments": [ { "argument_name": "create_notification_scheme", "argument_value": { "scheme_details": "Provide details for the new notification scheme." } }, { "argument_name": "update_notification_scheme", "argument_value": { "scheme_id": "LM10" } }, { "argument_name": "add_notifications_to_scheme", "argument_value": { "scheme_id": "LM10" } } ] } ] } Answer: [/INST] { "tools": [ { "tool_name": "issue_navigator_settings", "arguments": [ { "argument_name": "get_default_columns", "argument_value": "" } ] }, { "tool_name": "issue_notification_schemes", "arguments": [ { "argument_name": "create_notification_scheme", "argument_value": { "scheme_details": "Provide details for the new notification scheme." } }, { "argument_name": "update_notification_scheme", "argument_value": { "scheme_id": "LM10" } }, { "argument_name": "add_notifications_to_scheme", "argument_value": { "scheme_id": "LM10" } } ] } ] } </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Perform a comprehensive update on custom field options with id 789 and get details for a set of issues within a Forge app. APIs: [{'tool': 'issue_custom_field_values_(apps)', 'description': 'This resource represents the values of custom fields added by [Forge apps](https://developer.atlassian.com/platform/forge/). Use it to update the value of a custom field on issues.', 'arguments': [{'argument_name': 'app/field/value', 'argument description': 'Updates the value of one or more custom fields on one or more issues. Combinations of custom field and issue should be unique within the request. Custom fields can only be updated by the Forge app that created them.\n\n**[Permissions](#permissions) required:** Only the app that created the custom field can update its values with this operation.', 'argument type': 'application/json'}, {'argument_name': 'app/field/:fieldIdOrKey/value', 'argument description': 'Updates the value of a custom field on one or more issues. Custom fields can only be updated by the Forge app that created them.\n\n**[Permissions](#permissions) required:** Only the app that created the custom field can update its values with this operation.', 'argument type': 'application/json'}]}, {'tool': 'ui_modifications_(apps)', 'description': 'UI modifications is a feature available for **Forge apps only**. It enables Forge apps to control how selected Jira fields behave on the following views: global issue create, issue view. For example: hide specific fields, set them as required, etc.', 'arguments': [{'argument_name': 'uiModifications', 'argument description': 'Gets UI modifications. UI modifications can only be retrieved by Forge apps.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'uiModifications', 'argument description': 'Creates a UI modification. UI modification can only be created by Forge apps.\n\nEach app can define up to 3000 UI modifications. Each UI modification can define up to 1000 contexts. The same context can be assigned to maximum 100 UI modifications.\n\n**[Permissions](#permissions) required:**\n\n * *None* if the UI modification is created without contexts.\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for one or more projects, if the UI modification is created with contexts.', 'argument type': 'application/json'}, {'argument_name': 'uiModifications/:uiModificationId', 'argument description': 'Updates a UI modification. UI modification can only be updated by Forge apps.\n\nEach UI modification can define up to 1000 contexts. The same context can be assigned to maximum 100 UI modifications.\n\n**[Permissions](#permissions) required:**\n\n * *None* if the UI modification is created without contexts.\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for one or more projects, if the UI modification is created with contexts.', 'argument type': 'application/json'}, {'argument_name': 'uiModifications/:uiModificationId', 'argument description': 'Deletes a UI modification. All the contexts that belong to the UI modification are deleted too. UI modification can only be deleted by Forge apps.\n\n**[Permissions](#permissions) required:** None.', 'argument type': ''}]}, {'tool': 'issue_custom_field_configuration_(apps)', 'description': 'This resource represents configurations stored against a custom field context by a [Forge app](https://developer.atlassian.com/platform/forge/). Configurations are information used by the Forge app at runtime to determine how to handle or process the data in a custom field in a given context. Use this resource to set and read configurations.', 'arguments': [{'argument_name': 'app/field/:fieldIdOrKey/context/configuration', 'argument description': 'Returns a [paginated](#pagination) list of configurations for a custom field created by a [Forge app](https://developer.atlassian.com/platform/forge/).\n\nThe result can be filtered by one of these criteria:\n\n * `id`.\n * `fieldContextId`.\n * `issueId`.\n * `projectKeyOrId` and `issueTypeId`.\n\nOtherwise, all configurations are returned.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg). Jira permissions are not required for the Forge app that created the custom field.', 'argument type': 'application/json'}, {'argument_name': 'app/field/:fieldIdOrKey/context/configuration', 'argument description': 'Update the configuration for contexts of a custom field created by a [Forge app](https://developer.atlassian.com/platform/forge/).\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg). Jira permissions are not required for the Forge app that created the custom field.', 'argument type': 'application/json'}]}, {'tool': 'issue_fields', 'description': 'This resource represents issue fields, both system and custom fields. Use it to get fields, field configurations, and create custom fields.', 'arguments': [{'argument_name': 'field', 'argument description': 'Returns system and custom issue fields according to the following rules:\n\n * Fields that cannot be added to the issue navigator are always returned.\n * Fields that cannot be placed on an issue screen are always returned.\n * Fields that depend on global Jira settings are only returned if the setting is enabled. That is, timetracking fields, subtasks, votes, and watches.\n * For all other fields, this operation only returns the fields that the user has permission to view (that is, the field is used in at least one project that the user has *Browse Projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for.)\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'field', 'argument description': 'Creates a custom field.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/search', 'argument description': 'Returns a [paginated](#pagination) list of fields for Classic Jira projects. The list can include:\n\n * all fields\n * specific fields, by defining `id`\n * fields that contain a string in the field name or description, by defining `query`\n * specific fields that contain a string in the field name or description, by defining `id` and `query`\n\nOnly custom fields can be queried, `type` must be set to `custom`.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/search/trashed', 'argument description': 'Returns a [paginated](#pagination) list of fields in the trash. The list may be restricted to fields whose field name or description partially match a string.\n\nOnly custom fields can be queried, `type` must be set to `custom`.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:fieldId', 'argument description': 'Updates a custom field.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:fieldId/contexts', 'argument description': 'Returns a [paginated](#pagination) list of the contexts a field is used in. Deprecated, use [ Get custom field contexts](#api-rest-api-3-field-fieldId-context-get).\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:id', 'argument description': 'Deletes a custom field. The custom field is deleted whether it is in the trash or not. See [Edit or delete a custom field](https://confluence.atlassian.com/x/Z44fOw) for more information on trashing and deleting custom fields.\n\nThis operation is [asynchronous](#async). Follow the `location` link in the response to determine the status of the task and use [Get task](#api-rest-api-3-task-taskId-get) to obtain subsequent updates.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': ''}, {'argument_name': 'field/:id/restore', 'argument description': 'Restores a custom field from trash. See [Edit or delete a custom field](https://confluence.atlassian.com/x/Z44fOw) for more information on trashing and deleting custom fields.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:id/trash', 'argument description': 'Moves a custom field to trash. See [Edit or delete a custom field](https://confluence.atlassian.com/x/Z44fOw) for more information on trashing and deleting custom fields.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}]}] Answer: [/INST] { "APIs": [ { "tool": "issue_custom_field_values_(apps)", "arguments": [ { "argument_name": "app/field/value", "argument_value": "789" } ] }, { "tool": "issue_fields", "arguments": [ { "argument_name": "field", "argument_value": "custom field options" } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'Query': 'Get all work items similar to TKT-123, summarize them, create issues from that summary, and prioritize them', 'Solution': [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'TKT-123'}]}, {'tool_name': 'create_actionable_tasks_from_text', 'arguments': [{'argument_name': 'text', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'List all high severity tickets coming in from slack from customer Cust123 and generate a summary of them.', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'Cust123'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}, {'argument_name': 'ticket.severity', 'argument_value': 'high'}, {'argument_name': 'ticket.source_channel', 'argument_value': 'slack'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'Summarize high severity tickets from the customer UltimateCustomer', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'UltimateCustomer'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}] Current Query : Get all work items similar to TKT-456, summarize them, create issues from that summary, and prioritize them Current Solution : { "tools": [ { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-456" } }, { "tool": "summarize_objects", "arguments": { "objects": "$PREV[0]" } }, { "tool": "works_list", "arguments": { "applies_to_part": "$PREV[1]", "created_by": [], "issue.priority": [], "issue.rev_orgs": [], "limit": 50, "owned_by": [], "stage.name": [], "ticket.needs_response": false, "ticket.rev_org": [], "ticket.severity": [], "ticket.source_channel": [], "type": [] } }, { "tool": "add_work_items_to_sprint", "arguments": { "work_ids": "$PREV[2]", "sprint_id": "" } } ] } Answer: [/INST] { "tools": [ { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-123" } }, { "tool": "summarize_objects", "arguments": { "objects": "$PREV[0]" } }, { "tool": "create_actionable_tasks_from_text", "arguments": { "text": "$PREV[1]" } }, { "tool": "prioritize_objects", "arguments": { "objects": "$PREV[2]" } } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [[{'query': 'Create an enterprise authentication connection, retrieve its details, and then delete the connection.', 'solutions': [{'tool_name': 'dev-orgs.auth-connections.create', 'arguments': []}, {'tool_name': 'dev-orgs.auth-connections.get', 'arguments': [{'argument_name': 'id', 'argument_type': '$$PREV[0]'}]}, {'tool_name': 'dev-orgs.auth-connections.delete', 'arguments': []}]}], [{'query': 'Retrieve details of 123456 authentication connection and delete it.', 'solutions': [{'tool_name': 'dev-orgs.auth-connections.get', 'arguments': [{'argument_name': 'id', 'argument_value': '123456'}]}, {'tool_name': 'dev-orgs.auth-connections.delete', 'arguments': [{'argument_name': 'id', 'argument_value': '123456'}]}]}], [{'query': 'Create a JWT token for authentication, get its metadata, and then revoke the token.', 'solutions': [{'tool_name': 'auth-tokens.create', 'arguments': []}, {'tool_name': 'auth-tokens.get', 'arguments': [{'argument_name': 'token_id', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'auth-tokens.delete', 'arguments': []}]}]] Current Query : Create an enterprise authentication connection, retrieve its details, and then delete the connection. Current Solution : { "tools": [ { "tool": "dev-orgs.auth-connections.create", "arguments": [] }, { "tool": "dev-orgs.auth-connections.get", "arguments": [ { "argument_name": "id", "argument_value": "$PREV[0]" } ] }, { "tool": "dev-orgs.auth-connections.delete", "arguments": [ { "argument_name": "id", "argument_value": "$PREV[1]" } ] } ] } Answer: [/INST] { "tools": [ { "tool": "dev-orgs.auth-connections.create", "arguments": [] }, { "tool": "dev-orgs.auth-connections.get", "arguments": [ { "argument_name": "id", "argument_value": "$PREV[0]" } ] }, { "tool": "dev-orgs.auth-connections.delete", "arguments": [] } ] } </s>

<s>[INST] Given a query, solution and available APIs, optimize the number of APIs calls in the solution. Do not use API calls unless absolutely necessary. Look for redundancy and check for mistakes in the combination of API calls. Return current solution if no changes are necessary. Current Query : Find all high severity tickets coming in from messenger from customer AbraKaDabra and generate a summary of them. Add the summary to my current sprint. Current Solution : [{‘tool_name’: ‘whoami’, ‘arguments’: [] },{‘tool_name’: 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'AbraKaDabra'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': [‘$$PREV[0]’]}, {'argument_name': 'ticket.severity', 'argument_value': [‘high’]}, {'argument_name': 'ticket.source_channel', 'argument_value': [‘messenger’]},{'argument_name': 'type', 'argument_value': [‘ticket’]}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': ‘$$PREV[1]'}]}]Available APIs : [{ "tool": "who_am_i", "description": "Returns the ID of the current user", "arguments": [ { "argument_name": "", "argument_description": "", "argument_type": "", "examples": [] } ] },{'tool': 'get_sprint_id', 'description': 'Returns the ID of the current sprint', 'arguments': [{'argument_name': '', 'argument_description': '', 'argument_type': '', 'examples': []}]}, {'tool': 'add_work_items_to_sprint', 'description': 'Adds the given work items to the sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_description': 'A list of work item IDs to be added to the sprint.', 'argument_type': 'array of strings', 'examples': []}, {'argument_name': 'sprint_id', 'argument_description': 'The ID of the sprint to which the work items should be added', 'argument_type': 'str', 'examples': []}]}, {'tool': 'summarize_objects', 'description': 'Summarizes a list of objects. The logic of how to summarize a particular object type is an internal implementation detail.', 'arguments': [{'argument_name': 'objects', 'argument_description': 'List of objects to summarize', 'argument_type': 'array of objects', 'examples': []}]}, {'tool': 'prioritize_objects', 'description': 'Summarizes a list of objects. The logic of how to summarize a particular object type is an internal implementation detail.', 'arguments': [{'argument_name': 'objects', 'argument_description': 'List of objects to summarize', 'argument_type': 'array of objects', 'examples': []}]}] Final solution (same format as current solution): [/INST] [{‘tool_name’: ‘whoami’, ‘arguments’: [] },{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'AbraKaDabra'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': [‘$$PREV[0]’]}, {'argument_name': 'ticket.severity', 'argument_value': [‘high’]}, {'argument_name': 'ticket.source_channel', 'argument_value': [‘messenger’]},{'argument_name': 'type', 'argument_value': [‘ticket’]}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}, {'tool_name': 'get_sprint_id', 'arguments': []}, {'tool_name': 'add_work_items_to_sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_value': '$$PREV[2]'}, {'argument_name': 'sprint_id', 'argument_value': '$$PREV[3]'}]}] </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Retrieve default columns for the issue navigator, manage notification schemes for scheme id LM10(get existing, create new, update, and customize), and update default columns if necessary. APIs: [{'tool': 'issue_navigator_settings', 'description': 'This resource represents issue navigator settings. Use it to get and set issue navigator default columns.', 'arguments': [{'argument_name': 'settings/columns', 'argument description': 'Returns the default issue navigator columns.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'settings/columns', 'argument description': 'Sets the default issue navigator columns.\n\nThe `columns` parameter accepts a navigable field value and is expressed as HTML form data. To specify multiple columns, pass multiple `columns` parameters. For example, in curl:\n\n`curl -X PUT -d columns=summary -d columns=description https://your-domain.atlassian.net/rest/api/3/settings/columns`\n\nIf no column details are sent, then all default columns are removed.\n\nA navigable field is one that can be used as a column on the issue navigator. Find details of navigable issue columns using [Get fields](#api-rest-api-3-field-get).\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': '*/*'}]}, {'tool': 'ui_modifications_(apps)', 'description': 'UI modifications is a feature available for **Forge apps only**. It enables Forge apps to control how selected Jira fields behave on the following views: global issue create, issue view. For example: hide specific fields, set them as required, etc.', 'arguments': [{'argument_name': 'uiModifications', 'argument description': 'Gets UI modifications. UI modifications can only be retrieved by Forge apps.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'uiModifications', 'argument description': 'Creates a UI modification. UI modification can only be created by Forge apps.\n\nEach app can define up to 3000 UI modifications. Each UI modification can define up to 1000 contexts. The same context can be assigned to maximum 100 UI modifications.\n\n**[Permissions](#permissions) required:**\n\n * *None* if the UI modification is created without contexts.\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for one or more projects, if the UI modification is created with contexts.', 'argument type': 'application/json'}, {'argument_name': 'uiModifications/:uiModificationId', 'argument description': 'Updates a UI modification. UI modification can only be updated by Forge apps.\n\nEach UI modification can define up to 1000 contexts. The same context can be assigned to maximum 100 UI modifications.\n\n**[Permissions](#permissions) required:**\n\n * *None* if the UI modification is created without contexts.\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for one or more projects, if the UI modification is created with contexts.', 'argument type': 'application/json'}, {'argument_name': 'uiModifications/:uiModificationId', 'argument description': 'Deletes a UI modification. All the contexts that belong to the UI modification are deleted too. UI modification can only be deleted by Forge apps.\n\n**[Permissions](#permissions) required:** None.', 'argument type': ''}]}, {'tool': 'issue_priorities', 'description': 'This resource represents issue priorities. Use it to get, create and update issue priorities and details for individual issue priorities.', 'arguments': [{'argument_name': 'priority', 'argument description': 'Returns the list of all issue priorities.\n\n**[Permissions](#permissions) required:** Permission to access Jira.', 'argument type': 'application/json'}, {'argument_name': 'priority', 'argument description': 'Creates an issue priority.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'priority/default', 'argument description': 'Sets default issue priority.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'priority/move', 'argument description': 'Changes the order of issue priorities.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'priority/search', 'argument description': 'Returns a [paginated](#pagination) list of priorities. The list can contain all priorities or a subset determined by any combination of these criteria:\n\n * a list of priority IDs. Any invalid priority IDs are ignored.\n * a list of project IDs. Only priorities that are available in these projects will be returned. Any invalid project IDs are ignored.\n * whether the field configuration is a default. This returns priorities from company-managed (classic) projects only, as there is no concept of default priorities in team-managed projects.\n\n**[Permissions](#permissions) required:** Permission to access Jira.', 'argument type': 'application/json'}, {'argument_name': 'priority/:id', 'argument description': 'Returns an issue priority.\n\n**[Permissions](#permissions) required:** Permission to access Jira.', 'argument type': 'application/json'}, {'argument_name': 'priority/:id', 'argument description': 'Updates an issue priority.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'priority/:id', 'argument description': '*Deprecated: please refer to the* [changelog](https://developer.atlassian.com/changelog/#CHANGE-1066) *for more details.*\n\nDeletes an issue priority.\n\nThis operation is [asynchronous](#async). Follow the `location` link in the response to determine the status of the task and use [Get task](#api-rest-api-3-task-taskId-get) to obtain subsequent updates.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': ''}]}, {'tool': 'issue_fields', 'description': 'This resource represents issue fields, both system and custom fields. Use it to get fields, field configurations, and create custom fields.', 'arguments': [{'argument_name': 'field', 'argument description': 'Returns system and custom issue fields according to the following rules:\n\n * Fields that cannot be added to the issue navigator are always returned.\n * Fields that cannot be placed on an issue screen are always returned.\n * Fields that depend on global Jira settings are only returned if the setting is enabled. That is, timetracking fields, subtasks, votes, and watches.\n * For all other fields, this operation only returns the fields that the user has permission to view (that is, the field is used in at least one project that the user has *Browse Projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for.)\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'field', 'argument description': 'Creates a custom field.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/search', 'argument description': 'Returns a [paginated](#pagination) list of fields for Classic Jira projects. The list can include:\n\n * all fields\n * specific fields, by defining `id`\n * fields that contain a string in the field name or description, by defining `query`\n * specific fields that contain a string in the field name or description, by defining `id` and `query`\n\nOnly custom fields can be queried, `type` must be set to `custom`.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/search/trashed', 'argument description': 'Returns a [paginated](#pagination) list of fields in the trash. The list may be restricted to fields whose field name or description partially match a string.\n\nOnly custom fields can be queried, `type` must be set to `custom`.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:fieldId', 'argument description': 'Updates a custom field.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:fieldId/contexts', 'argument description': 'Returns a [paginated](#pagination) list of the contexts a field is used in. Deprecated, use [ Get custom field contexts](#api-rest-api-3-field-fieldId-context-get).\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:id', 'argument description': 'Deletes a custom field. The custom field is deleted whether it is in the trash or not. See [Edit or delete a custom field](https://confluence.atlassian.com/x/Z44fOw) for more information on trashing and deleting custom fields.\n\nThis operation is [asynchronous](#async). Follow the `location` link in the response to determine the status of the task and use [Get task](#api-rest-api-3-task-taskId-get) to obtain subsequent updates.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': ''}, {'argument_name': 'field/:id/restore', 'argument description': 'Restores a custom field from trash. See [Edit or delete a custom field](https://confluence.atlassian.com/x/Z44fOw) for more information on trashing and deleting custom fields.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'field/:id/trash', 'argument description': 'Moves a custom field to trash. See [Edit or delete a custom field](https://confluence.atlassian.com/x/Z44fOw) for more information on trashing and deleting custom fields.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}]}] Answer: [/INST] { "tools": [ { "tool": "issue_navigator_settings", "arguments": [ { "argument_name": "settings/columns", "argument_value": "default" } ] }, { "tool": "ui_modifications_(apps)", "arguments": [ { "argument_name": "uiModifications", "argument_value": "get existing" } ] }, { "tool": "issue_priorities", "arguments": [ { "argument_name": "priority", "argument_value": "default" } ] }, { "tool": "issue_fields", "arguments": [ { "argument_name": "field", "argument_value": "system and custom" } ] } ] } </s>

<s>[INST] Given a query, solution and available APIs, optimize the number of APIs calls in the solution. Do not use API calls unless absolutely necessary. Look for redundancy and check for mistakes in the combination of API calls. Return current solution if no changes are necessary. Current Query : Create a new part, retrieve its information, and list parts created by DEVU-12345. Current Solution : { "APIs": [ { "tool": "parts.create", "arguments": [] }, { "tool": "parts.get", "arguments": [ { "argument_name": "id", "argument_value": "$$PREV[0]" } ] }, { "tool": "parts.list", "arguments": [ { "argument_name": "created_by", "argument_value": ["DEVU-12345"] } ] } ] } Available APIs : [{'tool': 'parts.create', 'description': 'Creates new [part](https://devrev.ai/docs/product/parts).', 'arguments': []}, {'tool': 'parts.get', 'description': "Gets a [part's](https://devrev.ai/docs/product/parts) information.\n", 'arguments': [{'argument_name': 'id', 'argument_description': 'The ID of the part to retrieve.', 'argument_type': 'string', 'examples': 'PROD-12345'}]}, {'tool': 'parts.list', 'description': 'Lists a collection of [parts](https://devrev.ai/docs/product/parts).\n', 'arguments': [{'argument_name': 'created_by', 'argument_description': 'Filters for parts created by any of these users.', 'argument_type': 'array', 'examples': ['DEVU-12345']}, {'argument_name': 'cursor', 'argument_description': 'The cursor to resume iteration from. If not provided, then iteration\nstarts from the beginning.\n', 'argument_type': 'string', 'examples': []}, {'argument_name': 'limit', 'argument_description': "The maximum number of parts to return. The default is '50'.\n", 'argument_type': 'integer', 'examples': []}, {'argument_name': 'mode', 'argument_description': 'The iteration mode to use, otherwise if not set, then "after" is\nused.\n', 'argument_type': '', 'examples': []}, {'argument_name': 'name', 'argument_description': 'Filters for parts of the provided name(s).', 'argument_type': 'array', 'examples': []}, {'argument_name': 'owned_by', 'argument_description': 'Filters for parts owned by any of these users.', 'argument_type': 'array', 'examples': ['DEVU-12345']}, {'argument_name': 'parent_part.level', 'argument_description': 'Number of levels to fetch the part hierarchy up to.', 'argument_type': 'integer', 'examples': []}, {'argument_name': 'parent_part.parts', 'argument_description': 'Part IDs to fetch the hierarchy for. Required if any parent_part.*\nfields are provided.\n', 'argument_type': 'array', 'examples': ['PROD-12345']}, {'argument_name': 'type', 'argument_description': 'Filters for parts of the provided type(s).', 'argument_type': 'array', 'examples': []}]}, {'tool': 'parts.update', 'description': "Updates a [part's](https://devrev.ai/docs/product/parts) information.\n", 'arguments': []}] Final solution (same format as current solution): [/INST] { "APIs": [ { "tool": "parts.create", "arguments": [] }, { "tool": "parts.get", "arguments": [ { "argument_name": "id", "argument_value": "$$PREV[0]" } ] }, { "tool": "parts.list", "arguments": [ { "argument_name": "created_by", "argument_value": ["DEVU-12345"] } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'Query': 'Get all work items similar to TKT-123, summarize them, create issues from that summary, and prioritize them', 'Solution': [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'TKT-123'}]}, {'tool_name': 'create_actionable_tasks_from_text', 'arguments': [{'argument_name': 'text', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'List all high severity tickets coming in from slack from customer Cust123 and generate a summary of them.', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'Cust123'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}, {'argument_name': 'ticket.severity', 'argument_value': 'high'}, {'argument_name': 'ticket.source_channel', 'argument_value': 'slack'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'Summarize high severity tickets from the customer UltimateCustomer', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'UltimateCustomer'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}] Current Query : Get all work items similar to TKT-456, summarize them, create issues from that summary, and prioritize them Current Solution : { "tools": [ { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-123" } }, { "tool": "summarize_objects", "arguments": { "objects": "$PREV[0]" } }, { "tool": "create_actionable_tasks_from_text", "arguments": { "text": "$PREV[1]" } }, { "tool": "prioritize_objects", "arguments": { "objects": "$PREV[2]" } } ] } Answer: [/INST] [ { "tool_name":"get_similar_work_items", "arguments":[ { "argument_name": "work_id", "argument_value": "TKT-456" } ] }, { "tool_name": "create_actionable_tasks_from_text", "arguments": [ { "argument_name": "text", "argument_value": "$$PREV[0]" } ] }, { "tool_name": "prioritize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$$PREV[1]" } ] } ] } ] </s>

<s>[INST] Given a query, solution and available APIs, optimize the number of APIs calls in the solution. Do not use API calls unless absolutely necessary. Look for redundancy and check for mistakes in the combination of API calls. Return current solution if no changes are necessary. Current Query : Create a new timeline entry, retrieve the list of entries for ObjectID_Gamma_456, and update an entry. Current Solution : { "tools": [ { "tool_name": "timeline-entries.create", "arguments": [] }, { "tool_name": "timeline-entries.list", "arguments": [ { "argument_name": "object", "argument_value": "ObjectID_Gamma_456" } ] }, { "tool_name": "timeline-entries.update", "arguments": [] } ] } Available APIs : [{'tool': 'timeline-entries.create', 'description': "Creates a new entry on an object's timeline.", 'arguments': []}, {'tool': 'timeline-entries.update', 'description': "Updates an entry on an object's timeline.", 'arguments': []}, {'tool': 'timeline-entries.list', 'description': 'Lists the timeline entries for an object.', 'arguments': [{'argument_name': 'object', 'argument_description': 'The ID of the object to list timeline entries for.', 'argument_type': 'string', 'examples': 'PROD-12345'}, {'argument_name': 'cursor', 'argument_description': 'The cursor to resume iteration from. If not provided, then iteration\nstarts from the beginning.\n', 'argument_type': 'string', 'examples': []}, {'argument_name': 'limit', 'argument_description': 'The maximum number of entries to return. If not set, then this\ndefaults to `50`.\n', 'argument_type': 'integer', 'examples': []}, {'argument_name': 'mode', 'argument_description': 'The iteration mode to use, otherwise if not set, then "after" is\nused.\n', 'argument_type': '', 'examples': []}, {'argument_name': 'visibility', 'argument_description': 'The visibility of the timeline entries to filter for. Note this is a\nstrict filter, such that only entries with the exact visibilities\nspecified will be returned.\n', 'argument_type': 'array', 'examples': []}]}, {'tool': 'timeline-entries.get', 'description': "Gets an entry on an object's timeline.", 'arguments': [{'argument_name': 'id', 'argument_description': 'The ID of the timeline entry to get.', 'argument_type': 'string', 'examples': 'don:core:<partition>:devo/<dev-org-id>:ticket/123:timeline_event/<timeline-event-id>'}]}] Final solution (same format as current solution): [/INST] { "tools": [ { "tool_name": "timeline-entries.create", "arguments": [] }, { "tool_name": "timeline-entries.list", "arguments": [ { "argument_name": "object", "argument_value": "ObjectID_Gamma_456" } ] }, { "tool_name": "timeline-entries.update", "arguments": [] } ] } </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Summarize ideas that are similar to ABCD1234:devo/2:issue/32 APIs: [{'tool': 'summarize_objects', 'description': 'Summarizes a list of objects. The logic of how to summarize a particular object type is an internal implementation detail.', 'arguments': [{'argument_name': 'objects', 'argument_description': 'List of objects to summarize', 'argument_type': 'array of objects', 'examples': []}]}, {'tool': 'prioritize_objects', 'description': 'Summarizes a list of objects. The logic of how to summarize a particular object type is an internal implementation detail.', 'arguments': [{'argument_name': 'objects', 'argument_description': 'List of objects to summarize', 'argument_type': 'array of objects', 'examples': []}]}, {'tool': 'get_similar_work_items', 'description': 'Returns a list of work items that are similar to the given work item', 'arguments': [{'argument_name': 'work_id', 'argument_description': 'The ID of the work item for which you want to find similar items', 'argument_type': 'string', 'examples': []}]}, {'tool': 'get_sprint_id', 'description': 'Returns the ID of the current sprint', 'arguments': [{'argument_name': '', 'argument_description': '', 'argument_type': '', 'examples': []}]}] Answer: [/INST] { "tools": [ { "tool": "get_similar_work_items", "arguments": [ { "argument_name": "work_id", "argument_value": "ABCD1234:devo/2:issue/32" } ] }, { "tool": "summarize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$PREV[0]" } ] }, { "tool": "prioritize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$PREV[1]" } ] } ] } </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Retrieve all remote issue links for the issue ABCD1234, obtain a list of all issue resolution values, and search for issues using JQL APIs: [{'tool': 'issue_search', 'description': 'This resource represents various ways to search for issues. Use it to search for issues with a JQL query and find issues to populate an issue picker.', 'arguments': [{'argument_name': 'issue/picker', 'argument description': "Returns lists of issues matching a query string. Use this resource to provide auto-completion suggestions when the user is looking for an issue using a word or string.\n\nThis operation returns two lists:\n\n * `History Search` which includes issues from the user's history of created, edited, or viewed issues that contain the string in the `query` parameter.\n * `Current Search` which includes issues that match the JQL expression in `currentJQL` and contain the string in the `query` parameter.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** None.", 'argument type': 'application/json'}, {'argument_name': 'jql/match', 'argument description': 'Checks whether one or more issues would be returned by one or more JQL queries.\n\n**[Permissions](#permissions) required:** None, however, issues are only matched against JQL queries where the user has:\n\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project that the issue is in.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, issue-level security permission to view the issue.', 'argument type': 'application/json'}, {'argument_name': 'search', 'argument description': 'Searches for issues using [JQL](https://confluence.atlassian.com/x/egORLQ).\n\nIf the JQL query expression is too large to be encoded as a query parameter, use the [POST](#api-rest-api-3-search-post) version of this resource.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** Issues are included in the response where the user has:\n\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project containing the issue.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, issue-level security permission to view the issue.', 'argument type': 'application/json'}, {'argument_name': 'search', 'argument description': 'Searches for issues using [JQL](https://confluence.atlassian.com/x/egORLQ).\n\nThere is a [GET](#api-rest-api-3-search-get) version of this resource that can be used for smaller JQL query expressions.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** Issues are included in the response where the user has:\n\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project containing the issue.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, issue-level security permission to view the issue.', 'argument type': 'application/json'}]}, {'tool': 'issue_remote_links', 'description': "This resource represents remote issue links, a way of linking Jira to information in other systems. Use it to get, create, update, and delete remote issue links either by ID or global ID. The global ID provides a way of accessing remote issue links using information about the item's remote system host and remote system identifier.", 'arguments': [{'argument_name': 'issue/:issueIdOrKey/remotelink', 'argument description': 'Returns the remote issue links for an issue. When a remote issue link global ID is provided the record with that global ID is returned, otherwise all remote issue links are returned. Where a global ID includes reserved URL characters these must be escaped in the request. For example, pass `system=http://www.mycompany.com/support&id=1` as `system%3Dhttp%3A%2F%2Fwww.mycompany.com%2Fsupport%26id%3D1`.\n\nThis operation requires [issue linking to be active](https://confluence.atlassian.com/x/yoXKM).\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project that the issue is in.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, issue-level security permission to view the issue.', 'argument type': 'application/json'}, {'argument_name': 'issue/:issueIdOrKey/remotelink', 'argument description': 'Creates or updates a remote issue link for an issue.\n\nIf a `globalId` is provided and a remote issue link with that global ID is found it is updated. Any fields without values in the request are set to null. Otherwise, the remote issue link is created.\n\nThis operation requires [issue linking to be active](https://confluence.atlassian.com/x/yoXKM).\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * *Browse projects* and *Link issues* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project that the issue is in.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, issue-level security permission to view the issue.', 'argument type': 'application/json'}, {'argument_name': 'issue/:issueIdOrKey/remotelink', 'argument description': "Deletes the remote issue link from the issue using the link's global ID. Where the global ID includes reserved URL characters these must be escaped in the request. For example, pass `system=http://www.mycompany.com/support&id=1` as `system%3Dhttp%3A%2F%2Fwww.mycompany.com%2Fsupport%26id%3D1`.\n\nThis operation requires [issue linking to be active](https://confluence.atlassian.com/x/yoXKM).\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * *Browse projects* and *Link issues* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project that the issue is in.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is implemented, issue-level security permission to view the issue.", 'argument type': ''}, {'argument_name': 'issue/:issueIdOrKey/remotelink/:linkId', 'argument description': 'Returns a remote issue link for an issue.\n\nThis operation requires [issue linking to be active](https://confluence.atlassian.com/x/yoXKM).\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * *Browse projects* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project that the issue is in.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, issue-level security permission to view the issue.', 'argument type': 'application/json'}, {'argument_name': 'issue/:issueIdOrKey/remotelink/:linkId', 'argument description': 'Updates a remote issue link for an issue.\n\nNote: Fields without values in the request are set to null.\n\nThis operation requires [issue linking to be active](https://confluence.atlassian.com/x/yoXKM).\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * *Browse projects* and *Link issues* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project that the issue is in.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, issue-level security permission to view the issue.', 'argument type': 'application/json'}, {'argument_name': 'issue/:issueIdOrKey/remotelink/:linkId', 'argument description': 'Deletes a remote issue link from an issue.\n\nThis operation requires [issue linking to be active](https://confluence.atlassian.com/x/yoXKM).\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * *Browse projects*, *Edit issues*, and *Link issues* [project permission](https://confluence.atlassian.com/x/yodKLg) for the project that the issue is in.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, issue-level security permission to view the issue.', 'argument type': ''}]}, {'tool': 'issue_resolutions', 'description': 'This resource represents issue resolution values. Use it to obtain a list of all issue resolution values and the details of individual resolution values.', 'arguments': [{'argument_name': 'resolution', 'argument description': 'Returns a list of all issue resolution values.\n\n**[Permissions](#permissions) required:** Permission to access Jira.', 'argument type': 'application/json'}, {'argument_name': 'resolution', 'argument description': 'Creates an issue resolution.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'resolution/default', 'argument description': 'Sets default issue resolution.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'resolution/move', 'argument description': 'Changes the order of issue resolutions.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'resolution/search', 'argument description': 'Returns a [paginated](#pagination) list of resolutions. The list can contain all resolutions or a subset determined by any combination of these criteria:\n\n * a list of resolutions IDs.\n * whether the field configuration is a default. This returns resolutions from company-managed (classic) projects only, as there is no concept of default resolutions in team-managed projects.\n\n**[Permissions](#permissions) required:** Permission to access Jira.', 'argument type': 'application/json'}, {'argument_name': 'resolution/:id', 'argument description': 'Returns an issue resolution value.\n\n**[Permissions](#permissions) required:** Permission to access Jira.', 'argument type': 'application/json'}, {'argument_name': 'resolution/:id', 'argument description': 'Updates an issue resolution.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'resolution/:id', 'argument description': 'Deletes an issue resolution.\n\nThis operation is [asynchronous](#async). Follow the `location` link in the response to determine the status of the task and use [Get task](#api-rest-api-3-task-taskId-get) to obtain subsequent updates.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': ''}]}, {'tool': 'issue_links', 'description': 'This resource represents links between issues. Use it to get, create, and delete links between issues.\n\nTo use it, the site must have [issue linking](https://confluence.atlassian.com/x/yoXKM) enabled.', 'arguments': [{'argument_name': 'issueLink', 'argument description': 'Creates a link between two issues. Use this operation to indicate a relationship between two issues and optionally add a comment to the from (outward) issue. To use this resource the site must have [Issue Linking](https://confluence.atlassian.com/x/yoXKM) enabled.\n\nThis resource returns nothing on the creation of an issue link. To obtain the ID of the issue link, use `https://your-domain.atlassian.net/rest/api/3/issue/[linked issue key]?fields=issuelinks`.\n\nIf the link request duplicates a link, the response indicates that the issue link was created. If the request included a comment, the comment is added.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * *Browse project* [project permission](https://confluence.atlassian.com/x/yodKLg) for all the projects containing the issues to be linked,\n * *Link issues* [project permission](https://confluence.atlassian.com/x/yodKLg) on the project containing the from (outward) issue,\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, issue-level security permission to view the issue.\n * If the comment has visibility restrictions, belongs to the group or has the role visibility is restricted to.', 'argument type': 'application/json'}, {'argument_name': 'issueLink/:linkId', 'argument description': 'Returns an issue link.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * *Browse project* [project permission](https://confluence.atlassian.com/x/yodKLg) for all the projects containing the linked issues.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, permission to view both of the issues.', 'argument type': 'application/json'}, {'argument_name': 'issueLink/:linkId', 'argument description': 'Deletes an issue link.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:**\n\n * Browse project [project permission](https://confluence.atlassian.com/x/yodKLg) for all the projects containing the issues in the link.\n * *Link issues* [project permission](https://confluence.atlassian.com/x/yodKLg) for at least one of the projects containing issues in the link.\n * If [issue-level security](https://confluence.atlassian.com/x/J4lKLg) is configured, permission to view both of the issues.', 'argument type': ''}]}] Answer: [/INST] { "tools": [ { "tool": "issue_remote_links", "arguments": [ { "argument_name": "issue/:issueIdOrKey/remotelink", "argument_value": "ABCD1234" } ] }, { "tool": "issue_resolutions", "arguments": [ { "argument_name": "resolution", "argument_value": "" } ] }, { "tool": "issue_search", "arguments": [ { "argument_name": "search", "argument_value": "resolution = $PREV[1]" } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'Query': 'Prioritize my P0 issues and add them to the current sprint', 'Solution': [{'tool_name': 'who_am_i', 'arguments': []}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'issue.priority', 'argument_value': 'p0'}, {'argument_name': 'owned_by', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}, {'tool_name': 'get_sprint_id', 'arguments': []}, {'tool_name': 'add_work_items_to_sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_value': '$$PREV[2]'}, {'argument_name': 'sprint_id', 'argument_value': '$$PREV[3]'}]}]}, {'Query': 'Get all work items similar to TKT-123, summarize them, create issues from that summary, and prioritize them', 'Solution': [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'TKT-123'}]}, {'tool_name': 'create_actionable_tasks_from_text', 'arguments': [{'argument_name': 'text', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'What are my all issues in the triage stage under part FEAT-123? Summarize them.', 'Solution': [{'tool_name': 'who_am_i', 'arguments': []}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'stage.name', 'argument_value': 'triage'}, {'argument_name': 'applies_to_part', 'argument_value': 'FEAT-123'}, {'argument_name': 'owned_by', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}] Current Query : Prioritize my P0 issues and add them to the current sprint Current Solution : { "tools": [ { "tool": "get_sprint_id", "arguments": [] }, { "tool": "works_list", "arguments": [ { "argument_name": "issue.priority", "argument_value": "p0" } ] }, { "tool": "add_work_items_to_sprint", "arguments": [ { "argument_name": "work_ids", "argument_value": "$PREV[1]" }, { "argument_name": "sprint_id", "argument_value": "$PREV[0]" } ] } ] } Answer: [/INST] { "tools": [ { "tool": "who_am_i", "arguments": [] }, { "tool": "works_list", "arguments": [ { "argument_name": "issue.priority", "argument_value": "p0" }, { "argument_name": "owned_by", "argument_value": "$$PREV[0]" } ] }, { "tool": "prioritize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$$PREV[1]" } ] }, { "tool": "get_sprint_id", "arguments": [] }, { "tool": "add_work_items_to_sprint", "arguments": [ { "argument_name": "work_ids", "argument_value": "$$PREV[2]" }, { "argument_name": "sprint_id", "argument_value": "$$PREV[3]" } ] } ] } </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Retrieve tokens issued for the application 'app_456' and update token metadata for a specific token issued under a Dev organization. APIs: [{'tool': 'auth-tokens.update', 'description': 'Updates token metadata of a token issued under a given Dev\norganization.\n', 'arguments': []}, {'tool': 'auth-tokens.delete', 'description': 'Revokes the token that matches the given token ID issued under the\ngiven Dev organization.\n', 'arguments': []}, {'tool': 'auth-tokens.get', 'description': 'Gets the token metadata corresponding to the given token ID under the\ngiven Dev organization.\n', 'arguments': [{'argument_name': 'token_id', 'argument_description': 'The unique identifier of the token under a given Dev organization.\n', 'argument_type': 'string', 'examples': []}]}, {'tool': 'rev-orgs.create', 'description': "Creates a Rev organization in the authenticated user's Dev\norganization.\n", 'arguments': []}] Answer: [/INST] { "tools": [ { "tool": "auth-tokens.get", "arguments": [ { "argument_name": "token_id", "argument_value": "$PREV[0]" } ] }, { "tool": "auth-tokens.update", "arguments": [] } ] } </s>

<s>[INST] Given a query, solution and available APIs, optimize the number of APIs calls in the solution. Do not use API calls unless absolutely necessary. Look for redundancy and check for mistakes in the combination of API calls. Return current solution if no changes are necessary. Current Query : Retrieve the current announcement banner configuration including required permissions for access, and then get details of all application roles in Jira. Current Solution : { "APIs": [ { "tool": "announcement_banner", "arguments": [ { "argument_name": "include_permissions", "argument_value": "true" } ] }, { "tool": "application_roles", "arguments": [ { "argument_name": "applicationrole", "argument_value": "Jira" } ] } ] } Available APIs : [{'tool': 'announcement_banner', 'description': 'This resource represents an announcement banner. Use it to retrieve and update banner configuration.', 'arguments': [{'argument_name': 'announcementBanner', 'argument description': 'Returns the current announcement banner configuration.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'announcementBanner', 'argument description': 'Updates the announcement banner configuration.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}]}, {'tool': 'server_info', 'description': 'This resource provides information about the Jira instance.', 'arguments': [{'argument_name': 'serverInfo', 'argument description': 'Returns information about the Jira instance.\n\nThis operation can be accessed anonymously.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}]}, {'tool': 'application_roles', 'description': 'This resource represents application roles. Use it to get details of an application role or all application roles.', 'arguments': [{'argument_name': 'applicationrole', 'argument description': 'Returns all application roles. In Jira, application roles are managed using the [Application access configuration](https://confluence.atlassian.com/x/3YxjL) page.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'applicationrole/:key', 'argument description': 'Returns an application role.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}]}, {'tool': 'license_metrics', 'description': 'This resource represents license metrics. Use it to get available metrics for Jira licences.', 'arguments': [{'argument_name': 'instance/license', 'argument description': 'Returns licensing information about the Jira instance.\n\n**[Permissions](#permissions) required:** None.', 'argument type': 'application/json'}, {'argument_name': 'license/approximateLicenseCount', 'argument description': 'Returns the approximate number of user accounts across all Jira licenses. Note that this information is cached with a 7-day lifecycle and could be stale at the time of call.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}, {'argument_name': 'license/approximateLicenseCount/product/:applicationKey', 'argument description': 'Returns the total approximate number of user accounts for a single Jira license. Note that this information is cached with a 7-day lifecycle and could be stale at the time of call.\n\n**[Permissions](#permissions) required:** *Administer Jira* [global permission](https://confluence.atlassian.com/x/x4dKLg).', 'argument type': 'application/json'}]}] Final solution (same format as current solution): [/INST] { "APIs": [ { "tool": "announcement_banner", "arguments": [ { "argument_name": "include_permissions", "argument_value": "true" } ] }, { "tool": "application_roles", "arguments": [ { "argument_name": "applicationrole", "argument_value": "Jira" } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'Query': 'Prioritize my P0 issues and add them to the current sprint', 'Solution': [{'tool_name': 'who_am_i', 'arguments': []}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'issue.priority', 'argument_value': 'p0'}, {'argument_name': 'owned_by', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}, {'tool_name': 'get_sprint_id', 'arguments': []}, {'tool_name': 'add_work_items_to_sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_value': '$$PREV[2]'}, {'argument_name': 'sprint_id', 'argument_value': '$$PREV[3]'}]}]}, {'Query': 'Get all work items similar to TKT-123, summarize them, create issues from that summary, and prioritize them', 'Solution': [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'TKT-123'}]}, {'tool_name': 'create_actionable_tasks_from_text', 'arguments': [{'argument_name': 'text', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'List all high severity tickets coming in from slack from customer Cust123 and generate a summary of them.', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'Cust123'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}, {'argument_name': 'ticket.severity', 'argument_value': 'high'}, {'argument_name': 'ticket.source_channel', 'argument_value': 'slack'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}] Current Query : Given a customer meeting transcript T, create action items and add them to my current sprint Current Solution : [{'tool_name': 'get_sprint_id', 'arguments': []}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'created_by', 'argument_value': '$PREV[0]'}]}, {'tool_name': 'add_work_items_to_sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_value': '$PREV[1]'}, {'argument_name': 'sprint_id', 'argument_value': '$PREV[0]'}]}] Answer: [/INST] [ { "tool_name": "create_actionable_tasks_from_text", "arguments": [ { "argument_name": "text", "argument_value": "T" } ] }, { "tool_name": "get_sprint_id", "arguments": [] }, { "tool_name": "add_work_items_to_sprint", "arguments": [ { "argument_name": “work_ids", "argument_value": "$$PREV[0]" }, { "argument_name": "sprint_id", "argument_value": "$$PREV[1]" } ] } ] </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'Query': 'Get all work items similar to TKT-123, summarize them, create issues from that summary, and prioritize them', 'Solution': [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'TKT-123'}]}, {'tool_name': 'create_actionable_tasks_from_text', 'arguments': [{'argument_name': 'text', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'List all high severity tickets coming in from slack from customer Cust123 and generate a summary of them.', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'Cust123'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}, {'argument_name': 'ticket.severity', 'argument_value': 'high'}, {'argument_name': 'ticket.source_channel', 'argument_value': 'slack'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'Summarize high severity tickets from the customer UltimateCustomer', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'UltimateCustomer'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}] Current Query : Get all work items similar to TKT-456, TKT-458, T TKT-460, summarize them, create issues from that summary, and prioritize them Current Solution : { "tools": [ { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-456" } }, { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-458" } }, { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-460" } }, { "tool": "summarize_objects", "arguments": { "objects": [ "$PREV[0]", "$PREV[1]", "$PREV[2]" ] } }, { "tool": "works_list", "arguments": { "issue.priority": ["p0", "p1", "p2", "p3"] } }, { "tool": "add_work_items_to_sprint", "arguments": { "work_ids": ["$PREV[3]"], "sprint_id": "sprint_id" } } ] } Answer: [/INST] { "tools": [ { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-456" } }, { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-458" } }, { "tool": "get_similar_work_items", "arguments": { "work_id": "TKT-460" } }, { "tool": "summarize_objects", "arguments": { "objects": [ "$PREV[0]", "$PREV[1]", "$PREV[2]" ] } }, { "tool": "create_actionable_tasks_from_text", "arguments": { "text": "$PREV[3]" } }, { "tool": "prioritize_objects", "arguments": { "objects": [ "$PREV[4]" ] } } ] } </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Get details for part ID PROD-789, update its information, and delete the part with ID PROD-xyz123. APIs: [{'tool': 'parts.delete', 'description': 'Deletes a [part](https://devrev.ai/docs/product/parts).', 'arguments': []}, {'tool': 'parts.update', 'description': "Updates a [part's](https://devrev.ai/docs/product/parts) information.\n", 'arguments': []}, {'tool': 'parts.get', 'description': "Gets a [part's](https://devrev.ai/docs/product/parts) information.\n", 'arguments': [{'argument_name': 'id', 'argument_description': 'The ID of the part to retrieve.', 'argument_type': 'string', 'examples': 'PROD-12345'}]}, {'tool': 'works.update', 'description': "Updates a work item's information.", 'arguments': []}] Answer: [/INST] { "tools": [ { "tool": "parts.get", "arguments": [ { "argument_name": "id", "argument_value": "PROD-789" } ] }, { "tool": "parts.update", "arguments": [ { "argument_name": "id", "argument_value": "$PREV[0].id" } ] }, { "tool": "parts.delete", "arguments": [ { "argument_name": "id", "argument_value": "PROD-xyz123" } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [[{'query': "Retrieve all tokens for the subject 'user123' of a specific type and update the metadata of a token issued under a Dev organization.", 'solutions': [{'tool': 'auth-tokens.list', 'arguments': [{'argument_name': 'subject', 'argument_value': 'user123'}]}, {'tool': 'auth-tokens.update', 'arguments': []}]}], [{'query': "Retrieve tokens issued for the application 'app_456' and update token metadata for a specific token issued under a Dev organization.", 'solutions': [{'tool': 'auth-tokens.list', 'arguments': [{'argument_name': 'client_id', 'argument_value': 'app_456'}]}, {'tool': 'auth-tokens.update', 'arguments': []}]}], [{'query': "Retrieve all tokens issued for the application example_app and update the metadata for a token issued under the 'Dev' organization.", 'solutions': [{'tool': 'auth-tokens.list', 'arguments': [{'argument_name': 'client_id', 'argument_value': 'example_app'}]}, {'tool': 'auth-tokens.update', 'arguments': []}]}, {'query': 'Revoke all tokens of a specific type issued by the authenticated user.', 'solutions': [{'tool': 'auth-tokens.self.delete', 'arguments': []}]}]] Current Query : Retrieve all tokens for the subject 'user123' of a specific type and update the metadata of a token issued under a Dev organization.? Current Solution : { "tools": [ { "tool": "auth-tokens.list", "arguments": [ { "argument_name": "subject", "argument_value": "user123" } ] }, { "tool": "auth-tokens.update", "arguments": [] } ] } Answer: [/INST] { "tools": [ { "tool": "auth-tokens.list", "arguments": [ { "argument_name": "subject", "argument_value": "user123" } ] }, { "tool": "auth-tokens.update", "arguments": [] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'Query': 'What are my all issues in the triage stage under part FEAT-123? Summarize them.', 'Solution': [{'tool_name': 'who_am_i', 'arguments': []}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'stage.name', 'argument_value': 'triage'}, {'argument_name': 'applies_to_part', 'argument_value': 'FEAT-123'}, {'argument_name': 'owned_by', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'Summarize high severity tickets from the customer UltimateCustomer', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'UltimateCustomer'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'Prioritize my P0 issues and add them to the current sprint', 'Solution': [{'tool_name': 'who_am_i', 'arguments': []}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'issue.priority', 'argument_value': 'p0'}, {'argument_name': 'owned_by', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}, {'tool_name': 'get_sprint_id', 'arguments': []}, {'tool_name': 'add_work_items_to_sprint', 'arguments': [{'argument_name': 'work_ids', 'argument_value': '$$PREV[2]'}, {'argument_name': 'sprint_id', 'argument_value': '$$PREV[3]'}]}]}] Current Query : What are my all issues in the triage stage under part FEAT-123? Summarize them. Current Solution : { "tools": [ { "tool": "works_list", "arguments": [ { "argument_name": "applies_to_part", "argument_value": "FEAT-123" }, { "argument_name": "stage.name", "argument_value": "triage" } ] }, { "tool": "summarize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$PREV[0]" } ] } ] } Answer: [/INST] { "tools": [ { "tool": "who_am_i", "arguments": [] }, { "tool": "works_list", "arguments": [ { "argument_name": "stage.name", "argument_value": "triage" }, { "argument_name": "applies_to_part", "argument_value": "FEAT-123" }, { "argument_name": "owned_by", "argument_value": "$$PREV[0]" } ] }, { "tool": "summarize_objects", "arguments": [ { "argument_name": "objects", "argument_value": "$$PREV[1]" } ] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'query': 'Retrieve all remote issue links for the issue ABCD1234, obtain a list of all issue resolution values, and search for issues using JQL.', 'solutions': [{'tool_name': 'issue_remote_links', 'arguments': [{'argument_name': 'issue/:issueIdOrKey/remotelink', 'argument_value': 'ABCD1234'}]}, {'tool_name': 'issue_resolutions', 'arguments': [{'argument_name': 'resolution', 'argument_value': 'resolution_values'}]}, {'tool_name': 'issue_search', 'arguments': [{'argument_name': 'search', 'argument_value': 'JQL'}]}]}, {'query': 'Manage priorities(create and update)and properties for issues with priority id 123', 'solutions': [{'tool_name': 'issue_priorities', 'arguments': [{'argument_name': 'create_issue_priority', 'argument_value': {'priority_details': ''}}, {'argument_name': 'update_issue_priority', 'argument_value': {'priority_id': '123'}}, {'argument_name': 'priority/default', 'argument_value': {'default_priority': ''}}]}, {'tool_name': 'issue_properties', 'arguments': [{'argument_name': 'issue/properties/:propertyKey', 'argument_value': {'property_key': '', 'filter_criteria': ''}}]}]}, {'query': 'Retrieve default columns for the issue navigator, manage notification schemes for scheme id LM10(get existing, create new, update, and customize), and update default columns if necessary.', 'solution': [{'tool_name': 'issue_navigator_settings', 'arguments': [{'argument_name': 'get_default_columns', 'argument_value': ''}]}, {'tool_name': 'issue_notification_schemes', 'arguments': [{'argument_name': 'create_notification_scheme', 'argument_value': {'scheme_details': 'Provide details for the new notification scheme.'}}, {'argument_name': 'update_notification_scheme', 'argument_value': {'scheme_id': 'LM10'}}, {'argument_name': 'add_notifications_to_scheme', 'argument_value': {'scheme_id': 'LM10'}}]}]}] Current Query : What is meaning of life Current Solution : The APIs that can solve the query "What is meaning of life" are: 1. jira_expressions - expression/eval 2. issue_search - search To call the jira_expressions API, the following arguments need to be provided: - expression/eval To call the issue_search API, the following arguments need to be provided: - search The order in which the APIs need to be called is: 1. jira_expressions - expression/eval 2. issue_search - search Answer: [/INST] [] </s>

<s>[INST] Given a query and a list of APIs with their arguments, find whether the APIs can solve the query or not. If not, return empty list. Else, extract values or variables from the query that correspond to each argument of the tools that needs to be called. If some argument value isn't known, use APIs that can retrieve those argument values. Return a JSON file containing the tools in the order they need to be called. Identify cases where an API argument depends on the output of a previous API call. Replace such arguments with the notation $PREV[i] where 'i' represents the order of the previous API call in the chain. Query: Create a new timeline entry, retrieve the list of entries for ObjectID_Gamma_456, and update an entry. APIs: [{'tool': 'timeline-entries.create', 'description': "Creates a new entry on an object's timeline.", 'arguments': []}, {'tool': 'timeline-entries.update', 'description': "Updates an entry on an object's timeline.", 'arguments': []}, {'tool': 'timeline-entries.list', 'description': 'Lists the timeline entries for an object.', 'arguments': [{'argument_name': 'object', 'argument_description': 'The ID of the object to list timeline entries for.', 'argument_type': 'string', 'examples': 'PROD-12345'}, {'argument_name': 'cursor', 'argument_description': 'The cursor to resume iteration from. If not provided, then iteration\nstarts from the beginning.\n', 'argument_type': 'string', 'examples': []}, {'argument_name': 'limit', 'argument_description': 'The maximum number of entries to return. If not set, then this\ndefaults to `50`.\n', 'argument_type': 'integer', 'examples': []}, {'argument_name': 'mode', 'argument_description': 'The iteration mode to use, otherwise if not set, then "after" is\nused.\n', 'argument_type': '', 'examples': []}, {'argument_name': 'visibility', 'argument_description': 'The visibility of the timeline entries to filter for. Note this is a\nstrict filter, such that only entries with the exact visibilities\nspecified will be returned.\n', 'argument_type': 'array', 'examples': []}]}, {'tool': 'timeline-entries.get', 'description': "Gets an entry on an object's timeline.", 'arguments': [{'argument_name': 'id', 'argument_description': 'The ID of the timeline entry to get.', 'argument_type': 'string', 'examples': 'don:core:<partition>:devo/<dev-org-id>:ticket/123:timeline_event/<timeline-event-id>'}]}] Answer: [/INST] { "tools": [ { "tool": "timeline-entries.create", "arguments": [] }, { "tool": "timeline-entries.list", "arguments": [ { "argument_name": "object", "argument_description": "The ID of the object to list timeline entries for.", "argument_type": "string", "examples": "PROD-12345" }, { "argument_name": "cursor", "argument_description": "The cursor to resume iteration from. If not provided, then iteration\nstarts from the beginning.\n", "argument_type": "string", "examples": [] }, { "argument_name": "limit", "argument_description": "The maximum number of entries to return. If not set, then this\ndefaults to `50`.\n", "argument_type": "integer", "examples": [] }, { "argument_name": "mode", "argument_description": "The iteration mode to use, otherwise if not set, then \"after\" is\nused.\n", "argument_type": "", "examples": [] }, { "argument_name": "visibility", "argument_description": "The visibility of the timeline entries to filter for. Note this is a\nstrict filter, such that only entries with the exact visibilities\nspecified will be returned.\n", "argument_type": "array", "examples": [] } ] }, { "tool": "timeline-entries.update", "arguments": [] } ] } </s>

<s>[INST] Given a list of examples, each containing a query and the corresponding APIs to be called to solve the query, find whether the JSON present in current solution solves the current query. If yes, modify its format to match with that of solutions in examples and return the modified JSON. If not, modify the current JSON solution to include necessary tools from list of available tools. If no solution is available, return a blank list. Examples : [{'Query': 'List all high severity tickets coming in from slack from customer Cust123 and generate a summary of them.', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'Cust123'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}, {'argument_name': 'ticket.severity', 'argument_value': 'high'}, {'argument_name': 'ticket.source_channel', 'argument_value': 'slack'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'Summarize high severity tickets from the customer UltimateCustomer', 'Solution': [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'UltimateCustomer'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}, {'Query': 'Get all work items similar to TKT-123, summarize them, create issues from that summary, and prioritize them', 'Solution': [{'tool_name': 'get_similar_work_items', 'arguments': [{'argument_name': 'work_id', 'argument_value': 'TKT-123'}]}, {'tool_name': 'create_actionable_tasks_from_text', 'arguments': [{'argument_name': 'text', 'argument_value': '$$PREV[0]'}]}, {'tool_name': 'prioritize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]}] Current Query : Find all high severity tickets coming in from messenger from customer AbraKaDabra and generate a summary of them. Add the summary to my current sprint. Current Solution : { "APIs": [ { "tool": "get_sprint_id", "arguments": [] }, { "tool": "summarize_objects", "arguments": [ { "argument_name": "objects", "argument_description": "List of objects to summarize", "argument_type": "array of objects", "examples": [] } ] }, { "tool": "prioritize_objects", "arguments": [ { "argument_name": "objects", "argument_description": "List of objects to summarize", "argument_type": "array of objects", "examples": [] } ] }, { "tool": "add_work_items_to_sprint", "arguments": [ { "argument_name": "work_ids", "argument_description": "A list of work item IDs to be added to the sprint.", "argument_type": "array of strings", "examples": [] }, { "argument_name": "sprint_id", "argument_description": "The ID of the sprint to which the work items should be added", "argument_type": "str", "examples": [] } ] } ] } Answer: [/INST] [{'tool_name': 'search_object_by_name', 'arguments': [{'argument_name': 'query', 'argument_value': 'AbraKaDabra'}]}, {'tool_name': 'works_list', 'arguments': [{'argument_name': 'ticket.rev_org', 'argument_value': '$$PREV[0]'}, {'argument_name': 'ticket.severity', 'argument_value': 'high'}, {'argument_name': 'ticket.source_channel', 'argument_value': 'messenger'}]}, {'tool_name': 'summarize_objects', 'arguments': [{'argument_name': 'objects', 'argument_value': '$$PREV[1]'}]}]} </s>