tools_list stringlengths 173 352k | messages_history stringlengths 195 9.15k | binary_label stringclasses 2
values | label stringclasses 4
values | failure_reason stringclasses 632
values |
|---|---|---|---|---|
["{'name': 'add', 'description': 'Add two numbers', 'inputSchema': {'type': 'object', 'required': ['a', 'b'], 'properties': {'a': {'type': 'integer', 'title': 'A'}, 'b': {'type': 'integer', 'title': 'B'}}}, 'outputSchema': {'type': 'object', 'title': '_WrappedResult', 'required': ['result'], 'properties': {'result': {'... | [
{"role": "system", "content": "You are an assistant designed to manage employee-related tasks and information using a set of specialized tools. Here are the tools you have access to and how to use them:
* `add(event_name: str, start_date: str, end_date: str, attendees: list[str] = null, location: str = null)`:... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested 'Bob' to be an attendee, but the 'attendees' list in the incorrect parameters contains 'Bop' instead of 'Bob'. This is a one-letter typo that changes the intended attendee.'] |
["{'name': 'get_user_recent_tweets', 'description': '', 'inputSchema': {'type': 'object', 'title': 'get_user_recent_tweetsArguments', 'required': ['username'], 'properties': {'username': {'type': 'string', 'title': 'Username'}}}}"] | [
{
"role": "user",
"content": "Can you show me the last 10 tweets from Elon Musk?"
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_gz07qjwhpwqv1k0",
"... | INCORRECT | PARAM_VALUE_ERROR | ['The incorrect `username` 'elonmsuk' contains a subtle typo ('msuk' instead of 'musk'), which does not match the user's request for 'Elon Musk'. Additionally, the `count` of '11' is off by one from the requested '10' tweets.'] |
["{'type': 'function', 'function': {'name': 'retrieve_list_of_messages_in_user_chats', 'description': 'Retrieve list of messages in a user chats with parallel request', 'parameters': {'properties': {'user_chat_ids': {'type': 'array', 'items': {'type': 'string'}}}, 'required': ['user_chat_ids']}}}", "{'type': 'function'... | [
{
"role": "user",
"content": "Could you please retrieve the messages from the user chats with IDs 'uc_9876' and 'uc_5432'?"
},
{
"type": "function_call",
"id": "call_vfar4n7n8u8d9",
"name": "retrieve_list_of_messages_in_user_chats",
"arguments": {'user_... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'search_google', 'description': '\\n Perform a Google search and return formatted results.\\n \\n This function uses Google Custom Search API to search the web based on the provided query.\\n It formats the results into a consistent structure and handles potential errors.\\n \\n Args:\\n ... | [
{
"role": "user",
"content": "What's the weather like in London right now?"
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_g75m57l818dymn9dt9jyxbclhtxibb",
... | INCORRECT | PARAM_VALUE_ERROR | ['The incorrect query contains a subtle typo: 'currennt' instead of 'current'. This one-letter difference makes the parameter value invalid compared to the intended query.'] |
["{'name': 'search_pages', 'description': 'Search pages from Bookstack', 'inputSchema': {'type': 'object', 'properties': {'page': {'type': 'number', 'default': 1, 'minimum': 1, 'description': 'Page number to return'}, 'count': {'type': 'number', 'default': 10, 'maximum': 30, 'minimum': 1, 'description': 'Number of page... | [
{"role": "system", "content": "You are a helpful assistant with access to specialized tools. Use these tools to provide accurate and relevant information to the user.
Available tools:
- `search_pages(query: str)`: Searches the internet for general information. Use this when the user's query requires broad web kn... | INCORRECT | PARAM_VALUE_ERROR | ['The 'check_in_date' parameter is '2024-08-11' in the incorrect parameters, which is a one-day difference from the user's stated arrival date of 'the 10th' (2024-08-10).'] |
["{'type': 'function', 'function': {'name': 'arvix_survey', 'description': 'Get arvix paper related work and bibliography', 'parameters': {'properties': {'arvix_id': {'type': 'string', 'title': 'Arvix Id'}}, 'required': ['arvix_id']}}}"] | [
{
"role": "user",
"content": "Please get the related work and bibliography for the ArXiv paper with ID '2109.10852'."
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
["{'name': 'can_x_be_done', 'description': 'Check if a specific task can be done in a given language/framework/tool', 'inputSchema': {'type': 'object', 'required': ['docs', 'x', 'technology'], 'properties': {'x': {'type': 'string', 'description': 'The task to check'}, 'docs': {'type': 'string', 'description': 'Path to ... | [
{"role": "system", "content": "You are a helpful assistant with access to specialized tools. When a user asks a question, determine which tool is most appropriate to answer their request. Here are the tools you can use:
1. `can_x_be_done(task: str)`: Use this tool to inquire about the feasibility or possibility... | INCORRECT | PARAM_VALUE_ERROR | ['The parameter 'code_snippet' contains a minor typo ('variabl' instead of 'variable'), making it subtly different from the user's intended query about 'variable declarations'.'] |
["{'name': 'deploy_token', 'description': 'Deploy an ERC20 token contract', 'inputSchema': {'type': 'object', 'required': ['name', 'symbol', 'totalSupply'], 'properties': {'name': {'type': 'string', 'description': 'Token name'}, 'symbol': {'type': 'string', 'description': 'Token symbol'}, 'totalSupply': {'type': 'strin... | [
{
"role": "user",
"content": "I want to deploy a new ERC20 token called 'Super Coin' with the symbol 'SPC' and an initial supply of 1000000. It should have 18 decimals and be mintable."
},
{
"type": "function_call",
"id": "call_7h5myhfl2sc2ev5m5iz97o",
"name": ... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
['{\'name\': \'sell-token\', \'description\': \'Sell a Pump.fun token for SOL\', \'inputSchema\': {\'type\': \'object\', \'$schema\': \'http://json-schema.org/draft-07/schema#\', \'required\': [\'tokenAddress\', \'sellAmount\'], \'properties\': {\'sellAmount\': {\'type\': \'number\', \'description\': \'Amount of tokens... | [
{
"role": "user",
"content": "What is my balance for the token with address 9z4o1ZJ8e1yD6mXy2Kq4z3A5b6C7d8E9f0G1h2I3j4K5l6M7n8O9p in my trading_account?"
},
{
"id": "call_kqyjmr5vtket0brgo",
"type": "function",
"function": {
"name": "get-token-balan... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'xpath', 'description': 'Select query XML content using XPath', 'inputSchema': {'type': 'object', 'required': ['xml', 'query'], 'properties': {'xml': {'type': 'string', 'description': 'The XML content to query'}, 'query': {'type': 'string', 'description': 'The XPath query to execute'}, 'mimeType': {'type': '... | [
{
"role": "user",
"content": "I need to find patents related to 'quantum computing encryption' filed by 'IBM' in the US. Please show me the 25 newest patent applications filed after June 1st, 2022."
},
{
"type": "function_call",
"id": "call_1n1m1bdvcei8b7",
"na... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type', 'invalid enum value'] |
["{'type': 'function', 'function': {'name': 'getCommitHistory', 'description': 'Track the evolution of your Obsidian vault knowledge base by retrieving commit history from GitHub (string/string). See how your notes and ideas have developed over time with detailed diffs.', 'parameters': {'properties': {'days': {'type': ... | [
{"role": "system", "content": "You are an assistant with access to several tools. Use them to provide accurate information and complete tasks. Here's how you can use them:
- `getCommitHistory(repo_name: str, branch: str = "main", max_results: int = 10)`: Use this to retrieve the commit history for a given reposit... | INCORRECT | PARAM_VALUE_ERROR | ['The user query specified 'alice.smith', but the incorrect parameters provided 'alica.smith' which is a one-letter typo ('e' instead of 'a').'] |
['{\'name\': \'joinMeeting\', \'description\': \'Join a Zoom, Google Meet, or Microsoft Teams meeting using the provided meeting ID and password and capture transcript and audio recording\', \'inputSchema\': {\'type\': \'object\', \'$schema\': \'http://json-schema.org/draft-07/schema#\', \'required\': [\'platform\', \'... | [
{"role": "system", "content": "You are an assistant capable of managing meetings. You have access to the following tools:
- `joinMeeting(meetingId: string)`: Use this tool to join an ongoing or scheduled meeting. You must provide the exact meeting ID as a string. This is useful when the user wants to enter a spec... | INCORRECT | PARAM_VALUE_ERROR | ['The 'meetingId' in the incorrect parameters, 'conf-20231026-sales-reivew-007', contains a subtle typo: 'reivew' instead of the correct 'review'. The user's query clearly states 'review', making the 'reivew' parameter invalid.'] |
["{'name': 'list_directory', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['path'], 'properties': {'path': {'type': 'string', 'description': 'Directory path to list'}}, 'additionalProperties': False}}", "{'name': 'check_port', 'inputSchema': {'type': 'object', '$sc... | [
{"role": "system", "content": "You are an assistant with access to system utilities. To inspect directory contents, use the 'list_directory' tool. It requires a 'path' parameter, which is the absolute path to the directory you want to list. Use this when the user asks to see what files or folders are in a specifi... | INCORRECT | PARAM_VALUE_ERROR | ['The 'path' parameter in the incorrect call '/etc/nginxx' has an extra 'x' compared to the user's requested path '/etc/nginx'. This is a subtle one-letter difference that would lead to attempting to list a non-existent or incorrect directory.'] |
["{'name': 'channels_list_on_slack', 'description': 'List public channels in the workspace with pagination', 'inputSchema': {'type': 'object', 'properties': {'limit': {'type': 'number', 'default': 100, 'description': 'Maximum number of channels to return (default 100, max 200)'}, 'cursor': {'type': 'string', 'descripti... | [
{
"role": "user",
"content": "Can you fetch the profile information for the user with ID U023BECGF?"
},
{
"completion_message": {
"content": {
"type": "text",
"text": ""
},
"role": "assistant",
"st... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'update_collection', 'description': 'Update an existing collection in Outline', 'inputSchema': {'type': 'object', 'required': ['id'], 'properties': {'id': {'type': 'string', 'description': 'ID of the collection to update'}, 'name': {'type': 'string', 'description': 'New name of the collection'}, 'color': {'t... | [
{
"role": "user",
"content": "Please retrieve the document with the ID 'a1b2c3d4-e5f6-7890-1234-567890abcdef'."
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_muqe2... | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
["{'name': 'multiply', 'description': 'Multiplies two numbers.', 'inputSchema': {'type': 'object', 'title': 'MathOperation', 'required': ['a', 'b'], 'properties': {'a': {'type': 'integer', 'title': 'A', 'description': 'First number'}, 'b': {'type': 'integer', 'title': 'B', 'description': 'Second number'}}, 'description... | [
{"role": "system", "content": "You have access to the following tools: 'multiply' and 'add'.
- Use 'multiply(a, b)' to find the product of two numbers. This tool should be used when the user asks to multiply, find the product of, or scale one number by another. Both 'a' and 'b' must be numerical values.
- Use 'ad... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested the product of 'fifteen' and 'seven'. The incorrect parameters show 'b' as '8', which is an off-by-one error compared to the requested 'seven'.'] |
['{\'name\': \'nocodb-post-records\', \'description\': \'Nocodb - Post Records\', \'inputSchema\': {\'type\': \'object\', \'$schema\': \'http://json-schema.org/draft-07/schema#\', \'required\': [\'tableName\'], \'properties\': {\'data\': {\'description\': \'The data to be inserted into the table. \\n[WARNING] The struc... | [
{
"role": "user",
"content": "I need to add a new column for 'email' to the 'Customers' table. The data type should be 'SingleLineText'."
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'delete-silence', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['silenceId'], 'properties': {'silenceId': {'type': 'string', 'description': 'ID of the silence to delete'}}, 'additionalProperties': False}}", "{'name': 'get-alert-details', 'inputSchema': {... | [
{
"role": "user",
"content": "Can you show me the details for the alert with fingerprint 'a1b2c3d4e5f6g7h8'?"
},
{
"completion_message": {
"content": {
"type": "text",
"text": ""
},
"role": "assistant",
... | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
['{\'name\': \'speak\', \'description\': "Speak text aloud using Rime\'s text-to-speech API. Should be used when user asks you to speak or to announce and explain when you finish a command\\n \\nUser configuration:\\n\\nWHO_TO_ADDRESS: Developer\\n\\nWHEN_TO_SPEAK: when finishing a command\\n\\nVOICE: cove\\n\\nGUID... | [
{
"role": "user",
"content": "Could you please verbally tell me my current account balance?"
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_cz6r84r36kbl6baye3evwy0b... | INCORRECT | PARAM_VALUE_ERROR | ['The user asked for their 'current account balance', implying a specific factual amount. Changing 'five hundred dollars' to 'four hundred dollars' is a subtle numerical error that misrepresents the actual balance, making the spoken output factually incorrect despite being syntactically similar.'] |
["{'name': 'generate_dev_context', 'description': 'Generate AI summary and context for development work', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['topic'], 'properties': {'limit': {'type': 'number', 'default': 10, 'description': 'Number of documents to refere... | [
{"role": "system", "content": "You are an AI assistant designed to help users manage and access information within Notion. You have access to the following tools:
- `generate_dev_context(file_paths: list[str])`: Generates a development context from specified file paths. Use this when the user asks for context ab... | INCORRECT | PARAM_VALUE_ERROR | ['The `document_id` parameter has a subtle casing mismatch; the user requested '...j5k' but the parameter is '...j5K'. Additionally, the `num_lines` parameter is '11' instead of the requested '10', representing an off-by-one error.'] |
["{'name': 'mcpjam_search_mcp_spec', 'description': 'Search the MCP specification document for relevant content', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['query'], 'properties': {'query': {'enum': ['Introduction', 'Core components', 'Connection lifecycle', 'E... | [
{"role": "system", "content": "You are an AI assistant specialized in Stata data analysis. You have access to several tools to help manage and execute Stata tasks. When using these tools, always provide the full file name including the `.do` or `.log` extension where appropriate. Ensure do-file content is valid S... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
["{'name': 'search', 'description': '\\n Search DuckDuckGo and return formatted results.\\n\\n Args:\\n query: The search query string\\n max_results: Maximum number of results to return (default: 10)\\n ctx: MCP context for logging\\n ', 'inputSchema': {'type': 'object', 'title': 'searchA... | [
{
"role": "user",
"content": "Please find all project contexts associated with project ID 'quantum-leap-01' and tagged as 'final-report'."
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
... | INCORRECT | PARAM_VALUE_ERROR | ['invalid enum value', 'wrong data type'] |
["{'name': 'execute_swap', 'description': 'Execute a token swap', 'inputSchema': {'type': 'object', 'required': ['quote', 'walletPrivateKey'], 'properties': {'quote': {'type': 'object', 'description': 'Quote object from get_swap_quote'}, 'walletPrivateKey': {'type': 'string', 'description': 'Base58 encoded private key ... | [
{"role": "system", "content": "You are an AI assistant capable of generating speech and lipsync videos.
- Use the `speak(text: string, voice_id: string)` tool to generate only audio for a given text using a specified voice.
- Use the `create_lipsync_video_by_text(text: string, voice_id: string, avatar_id: strin... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'type': 'function', 'function': {'name': 'kibela_get_folder_notes', 'description': 'Get notes in a folder', 'parameters': {'properties': {'folderId': {'type': 'string', 'description': 'Folder ID'}}, 'required': ['folderId']}}}", '{\'type\': \'function\', \'function\': {\'name\': \'kibela_get_note_from_path\', \'desc... | [
{
"role": "user",
"content": "Can you retrieve the content of the Kibela note from this URL: 'https://myteam.kibe.la/notes/5154'?"
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
['{\'name\': \'get_editor_state\', \'tags\': [\'unity\', \'editor\', \'state\', \'hierarchy\', \'project\'], \'returns\': {\'type\': \'object\', \'format\': \'The response format varies based on the format parameter:\\n- Raw: Full UnityEditorState object\\n- scripts only: Array of script file paths\\n- no scripts: Unit... | [
{
"role": "user",
"content": "Could you please tell me the total number of API calls made in the past week?"
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_yiqu3",
... | INCORRECT | PARAM_VALUE_ERROR | ['The 'metric_name' parameter 'api_call' is a singular form, while the user query 'total number of API calls' implies a plural metric. The correct parameter should be 'api_calls' to match the plural usage in the query, indicating a count of multiple calls.'] |
["{'type': 'function', 'function': {'name': 'play_video', 'description': 'Play the video of the given stream link', 'parameters': {'properties': {'stream_link': {'type': 'string', 'title': 'Stream Link'}}, 'required': ['stream_link']}}}", "{'name': 'call_director', 'description': '$20', 'inputSchema': {'type': 'object'... | [
{"role": "system", "content": "You are an AI assistant designed to manage podcast content and interact with system functionalities. Your capabilities include managing podcast episodes (getting details, updating, publishing, deleting), handling user authentication (generating OAuth URLs, exchanging/refreshing toke... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'type': 'function', 'function': {'name': 'getLiveTemperature', 'description': '\\n Get live temperature for a specific location.\\n\\n Args:\\n latitude: The latitude coordinate (e.g., 40.7128 for New York)\\n longitude: The longitude coordinate (e.g., -74.0060 for New York)\\n\\n Returns:\\n ... | [
{"role": "system", "content": "You are an AI assistant with access to several tools. Use them judiciously to fulfill user requests.
- `getLiveTemperature(location: str)`: Use this tool to retrieve the current, real-time temperature for a specified geographic `location`. Only use this when the user explicitly asks... | INCORRECT | PARAM_VALUE_ERROR | ['The user specifically requested information 'for this year', which clearly implies 2024. The incorrect parameter uses '2023' instead of '2024', which is an off-by-one year mismatch and does not align with the user's current year request.'] |
['{\'name\': \'get_actions\', \'description\': \'Get all actions an actor can perform on a resource\', \'inputSchema\': {\'type\': \'object\', \'$schema\': \'http://json-schema.org/draft-07/schema#\', \'required\': [\'actorType\', \'actorId\', \'resourceType\', \'resourceId\'], \'properties\': {\'actorId\': {\'type\': ... | [
{"role": "system", "content": "You are an AI assistant with access to a `get_weather` tool. When the user asks for current weather information, use the `get_weather` tool. The tool requires a 'location' parameter to fetch the weather details. Always provide the full location including city and country if availabl... | INCORRECT | PARAM_VALUE_ERROR | ['invalid enum value', 'incorrect data type'] |
['{\'name\': \'get_history\', \'description\': \'\\nGet the history of an entity\\\'s state changes\\n\\nArgs:\\n entity_id: The entity ID to get history for\\n hours: Number of hours of history to retrieve (default: 24)\\n\\nReturns:\\n A dictionary containing:\\n - entity_id: The entity ID requested\\n ... | [
{
"role": "user",
"content": "What is the current status of the living room lamp?"
},
{
"id": "call_ugk1d85iaq9",
"type": "function",
"function": {
"name": "get_entity",
"arguments": {'entity_id': 'light.living_roon_lamp'}
}
... | INCORRECT | PARAM_VALUE_ERROR | ['The user query specifically refers to 'living room lamp', but the 'entity_id' in the incorrect parameters contains a subtle typo, using 'roon' instead of 'room', making it refer to a non-existent or incorrect entity.'] |
["{'type': 'function', 'function': {'name': 'get_last_topups', 'description': 'Get the last N top-up transactions', 'parameters': {'properties': {'limit': {'type': 'integer', 'title': 'Limit', 'default': 3}}, 'required': []}}}", "{'type': 'function', 'function': {'name': 'count_topups_by_number', 'description': 'Count ... | [
{
"role": "user",
"content": "Show me my last 5 top-up transactions."
},
{
"id": "call_odfa3ap450w5eq1r2z0br6i1",
"type": "function",
"function": {
"name": "get_last_topups",
"arguments": {'limit': 'five'}
}
}
] | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'list_insights', 'description': 'List all available Liquidium insights for the default project.\\n \\n Args:\\n search: Optional search query to filter insights\\n ', 'inputSchema': {'type': 'object', 'title': 'list_insightsArguments', 'properties': {'search': {'anyOf': [{'type': 'string'}, {... | [
{
"role": "user",
"content": "I need to see all the details for insight ID INS-2023-08-01-A123. Can you pull that up for me?"
},
{
"type": "function_call",
"id": "call_2u45co0u2",
"name": "insight_details",
"arguments": {'insight_id': 'INS-2023-08-01-A1... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested details for 'INS-2023-08-01-A123', but the 'insight_id' parameter in the incorrect tool call uses 'INS-2023-08-01-A122', which is a one-digit difference from the requested ID.'] |
["{'name': 'upload-file', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['filePath'], 'properties': {'filePath': {'type': 'string', 'description': 'è¦\\x81ä¸\\x8aä¼\\xa0ç\\x9a\\x84æ\\x96\\x87件路å¾\\x84'}}, 'additionalProperties': False}}"] | [
{"role": "system", "content": "You have access to the `upload-file` tool. Use this tool to upload files to specified directories. Call `upload-file` when the user clearly indicates an intention to upload a file, providing the `file_name`, the `destination_folder`, and whether to `overwrite_if_exists`. For example... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly specified the '2023 financial reports folder' in their request, but the `destination_folder` parameter in the incorrect tool call is '/financial/2024_reports/'. This is a subtle one-number difference (2023 vs 2024) that would cause the file to be uploaded to the wrong year's folder, deviating from... |
["{'type': 'function', 'function': {'name': 'amazon_sp_create_destination', 'description': 'Create a notification destination in Amazon Selling Partner API', 'parameters': {'properties': {'region': {'enum': ['na', 'eu', 'fe'], 'type': 'string'}, 'clientId': {'type': 'string'}, 'clientSecret': {'type': 'string'}, 'refre... | [
{
"role": "user",
"content": "I need to retrieve the details for a notification subscription from my Amazon Selling Partner account. My credentials are: client ID 'amzn1.application-oa2-client.xxxxxxxx', client secret 'secret_key_123', refresh token 'refresh_token_abc', and selling partner ID... | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
["{'type': 'function', 'function': {'name': 'execute_sql', 'description': 'Execute an SQL query on the MSSQL server', 'parameters': {'properties': {'query': {'type': 'string', 'description': 'The SQL query to execute'}}, 'required': ['query']}}}", '{\'name\': \'get_models\', \'description\': \'\\n Get all availa... | [
{
"role": "user",
"content": "Could you create a new voice for me? I'd like to name her Aria. She should be an adult female with an American English accent, speaking US English. I envision her as a clear, professional, and friendly voice, good for narrations."
},
{
"type": "fu... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested an 'American English accent' in their query, but the 'accent' parameter in the incorrect tool call is set to 'British English'. This is a subtle one-word difference that changes the intended characteristic of the voice.'] |
["{'type': 'function', 'function': {'name': 'get_page', 'description': 'Get a page from mediawiki.org\\nArgs:\\n title: The title of the page to get, which can be found in title field of the search results\\nReturns:\\n The page content\\n', 'parameters': {'properties': {'title': {'type': 'string', 'title': 'Titl... | [
{
"role": "user",
"content": "Could you please fetch the full content of the Wikipedia page for Artificial Intelligence?"
},
{
"completion_message": {
"content": {
"type": "text",
"text": ""
},
"role": "assist... | INCORRECT | PARAM_VALUE_ERROR | ['The 'incorrect_tool_call_1_params' has a subtle typo in the URL path. The correct spelling for 'intelligence' is used in the user's query and the correct parameters, but the incorrect parameters omit the second 'l' ('inteligence' instead of 'intelligence'). This single-letter difference would lead to a 404 or an inco... |
["{'name': 'web-search', 'description': 'Search the web', 'inputSchema': {'type': 'object', 'required': ['query'], 'properties': {'query': {'type': 'string', 'description': 'The search query.'}, 'byo_urls': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Bring your own URLs to search within.'}, 'ai_overv... | [
{"role": "system", "content": "You are a helpful assistant with access to two tools: 'web-search' and 'saveUrlContent'.
1. 'web-search': Use this tool to find information on the internet, answer questions about current events, facts, definitions, or anything that requires up-to-date or broad knowledge. Call this... | INCORRECT | PARAM_VALUE_ERROR | ['The 'url' in the incorrect parameters contains '2024' while the user specifically mentioned '2023' and provided a URL with '2023'. Additionally, the 'content_tag' 'AI Breakthru' has a subtle typo, missing the 'ough' from 'Breakthrough' as specified by the user.'] |
['{\'type\': \'function\', \'function\': {\'name\': \'fetch_public_market_trades\', \'description\': "Fetches recent public trades for a specific trading symbol. Does not require authentication, but providing API keys might increase rate limits or access. If fetching for non-spot markets (futures, options, swaps), ensu... | [
{
"role": "user",
"content": "Could you fetch the global long/short account ratio for the BTC/USDT futures market on Binance? I need the data for a 5-minute period. Please use the 'fapiPublicGetGlobalLongShortAccountRatio' method."
},
{
"role": "assistant",
"stop_reaso... | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
['{\'name\': \'search_code\', \'description\': "Fetches code that matches the provided regex pattern in `query`. This is NOT a semantic search.\\n Results are returned as an array of matching files, with the file\'s URL, repository, and language.\\n If you receive an error that indicates that you\'re not authenti... | [
{
"role": "user",
"content": "Show me examples of `express.Router()` being used in TypeScript files, and include code snippets."
},
{
"id": "call_eqn1m20o",
"type": "function",
"function": {
"name": "search_code",
"arguments": {'query': ... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
['{\'name\': \'multi_browserbase_stagehand_navigate_session\', \'description\': "Navigate to a URL in the browser. Only use this tool with URLs you\'re confident will work and stay up to date. Otherwise, use https://google.com as the starting point (for a specific session)", \'inputSchema\': {\'type\': \'object\', \'$s... | [
{
"role": "user",
"content": "I need to delete the development branch identified as 'feat-new-login-flow'. Can you do that for me?"
},
{
"type": "function_call",
"id": "call_lou524l",
"name": "delete_branch",
"arguments": {'branch_id': 12345}
}
] | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
["{'type': 'function', 'function': {'name': 'get_credit_card_transactions', 'description': 'Retrieve information about your Ramp credit card transactions.', 'parameters': {'properties': {'start': {'type': 'string'}, 'state': {'type': 'string'}, 'card_id': {'type': 'string', 'format': 'uuid'}, 'to_date': {'type': 'strin... | [
{"role": "system", "content": "You are a helpful assistant with access to a suite of tools. When a user asks a question, determine which tool or tools are appropriate to use, and call them with the correct parameters to fulfill the request. Your available tools include: get-funding-options, get-document-checklist... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
["{'name': 'decisionFramework', 'description': 'A detailed tool for structured decision analysis and rational choice.\\nThis tool helps models systematically evaluate options, criteria, and outcomes.\\nIt supports multiple decision frameworks, probability estimates, and value judgments.\\n\\nWhen to use this tool:\\n- ... | [
{"role": "system", "content": "You are an AI assistant equipped with a `decisionFramework` tool. Use this tool when the user expresses a need to make a choice, compare multiple options, or requires a structured analysis to arrive at a conclusion. The tool helps by evaluating different alternatives against specifi... | INCORRECT | PARAM_VALUE_ERROR | ['The 'decision_type' parameter contains a subtle typo, 'Selction' instead of 'Selection', which is a one-letter difference. Additionally, the 'criteria' parameter omits '/Libraries' from 'Ecosystem/Libraries', making 'Ecosystem' less specific and complete than what was requested by the user's mention of 'available lib... |
["{'type': 'function', 'function': {'name': 'query_collection', 'description': '\\n Search a Qdrant collection using semantic search with OpenAI embeddings.\\n\\n Args:\\n collection_name: Name of the Qdrant collection to search\\n query_text: The search query in natural language\\n limit: Ma... | [
{"role": "system", "content": "You are an AI assistant with access to tools. Use the `query_collection` tool to search for specific items within a named collection, providing a `query_text` and optionally `filters` and a `limit` on results. Use `collection_info` when you need to understand the structure or availa... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly asked for the 'top 5' papers, but the incorrect parameters set the 'limit' to 6, which is an off-by-one error and does not match the user's stated preference.'] |
['{\'type\': \'function\', \'function\': {\'name\': \'mmr_search\', \'description\': \'\\nSearch a PGVector collection using MMR (Maximal Marginal Relevance) search.\\n\\nThe connected vector database contains comprehensive restaurant information \\nfor Seoul, South Korea, including restaurant names, addresses, cuisine... | [
{
"role": "user",
"content": "Could you please retrieve the glucose levels for John Doe over the last three months? I'm only interested in the top 5 most recent results."
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
... | INCORRECT | PARAM_VALUE_ERROR | ['The `query` parameter contains a subtle typo ('Jon Doe' instead of 'John Doe'). The `date_range` parameter specifies 'last 2 months' which is a numerical difference from the user's request for 'last three months'. The `max_results` is 6, which is an off-by-one error compared to the user's request for 'top 5'. Additio... |
["{'name': 'extract_key_moments', 'description': 'Extracts key moments (timestamps and descriptions) from a given YouTube video.', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['youtube_url'], 'properties': {'youtube_url': {'type': 'string', 'format': 'uri'}, 'numb... | [
{"role": "system", "content": "You are an AI assistant capable of interacting with YouTube videos and SQL databases. You have access to the following tools:
- `extract_key_moments(video_url: str)`: Use this tool to identify and list significant timestamps or segments within a YouTube video. Call this when the us... | INCORRECT | PARAM_VALUE_ERROR | ['The 'video_url' in the incorrect parameters has a single character difference ('V' instead of 'U') in the video ID part (`YcZgK0fC2aV` vs `YcZgK0fC2aU`) compared to the URL provided in the user's query. This subtle change would lead to an incorrect or non-existent video being queried.'] |
["{'name': 'order_food', 'description': 'Order food from a restaurant.\\n \\n Args:\\n restaurant_url: URL of the restaurant\\n item_name: Name of the item to order\\n ', 'inputSchema': {'type': 'object', 'title': 'order_foodArguments', 'required': ['item_url', 'item_name'], 'properties': {'item_... | [
{
"role": "user",
"content": "Could you show me what Italian dessert options are available?"
},
{
"id": "call_ad9b9dou3tdye4ix910gofusu",
"type": "function",
"function": {
"name": "find_menu_options",
"arguments": {'cuisine': 'Italian', ... | INCORRECT | PARAM_VALUE_ERROR | ['The user asked for 'dessert' options, which implies a food category with two 's's. The 'category' parameter in the incorrect call is 'desert' (with one 's'), which is a typo and refers to an arid region, not a sweet course.'] |
['{\'name\': \'find_typescript_definition\', \'description\': \'Use /ts-def to trigger this tool. This tool can find the definition of a TypeScript symbol in your codebase. When you encounter an imported symbol (e.g., \\\'import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"\\\'), this tool w... | [
{"role": "system", "content": "You are an AI assistant designed to provide weather information. You have access to the following tools:
- `get_alerts()`: Retrieves any active weather alerts for your current region.
- `get_weather_for_location(location: string, unit: 'celsius' | 'fahrenheit')`: Provides the curren... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'type': 'function', 'function': {'name': 'slopwatch_status', 'description': 'Get current slop score and statistics', 'parameters': {'properties': {'random_string': {'type': 'string', 'description': 'Dummy parameter for no-parameter tools'}}, 'required': ['random_string']}}}", "{'type': 'function', 'function': {'name... | [
{
"role": "user",
"content": "Tell me about the geological time scale 250 million years ago."
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_5m4qjiu6",
... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
['{\'name\': \'get_papers_by_ids\', \'description\': "\\nRetrieves full details for a list of papers using their arXiv IDs from sources like arXiv.\\nArgs:\\n arxiv_ids: A list of arXiv IDs (e.g., [\'2307.09288\', \'1706.03762\']).\\n", \'inputSchema\': {\'type\': \'object\', \'title\': \'get_papers_by_idsArguments\... | [
{
"role": "user",
"content": "Could you fetch me the details for the paper identified as 2305.00123?"
},
{
"type": "function_call",
"id": "call_8ziq30aazup7lw0b0n1soiijvf6",
"name": "get_paper_by_id",
"arguments": {'paper_id': '2305.00124'}
}
] | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested information for paper ID '2305.00123'. The incorrect parameter '2305.00124' is off by a single digit, leading to a request for a different paper than what the user intended, making it an off-by-one error.'] |
["{'name': 'search_photos', 'description': '\\n Search for Unsplash photos\\n \\n Args:\\n query: Search keyword\\n page: Page number (1-based)\\n per_page: Results per page (1-30)\\n order_by: Sort method (relevant or latest)\\n color: Color filter (black_and_white, black, w... | [
{"role": "system", "content": "You are an AI assistant with access to a suite of tools to help users. Here's how you can use them:
1. `search_photos(query: str)`: Use this tool to find images based on a descriptive query. For example, if a user asks for 'pictures of cats' or 'photos of the Eiffel Tower'.
2. `sc... | INCORRECT | PARAM_VALUE_ERROR | ['The 'url' parameter in the incorrect tool call specifies '2024' as the year in the path, but the user's query explicitly references 'Q3 tech trends in 2023' and provides a URL with '2023'. The year '2024' is a subtle one-number difference that causes a mismatch with the user's stated intent and provided URL.'] |
["{'type': 'function', 'function': {'name': 'search_meal', 'description': '\\n Search for a meal by name and return cooking instructions.\\n ', 'parameters': {'properties': {'meal_name': {'type': 'string', 'title': 'Meal Name'}}, 'required': ['meal_name']}}}", "{'type': 'function', 'function': {'name': 'get_node'... | [
{"role": "system", "content": "You are a helpful assistant with access to several tools. Use them to assist the user in their tasks.
Available tools:
- `search_meal(query: str)`: Searches for meal recipes based on a query.
- `get_node(node_id: str)`: Retrieves detailed information about a specific node in a net... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested the rule for 'production authentication service', specified as 'auth-service-prod'. The incorrect parameters specify 'auth-service-stage', which is a different environment (staging instead of production).'] |
["{'name': 'list_database_servers', 'description': 'List all available database servers configured in the JDBCX server.\\n\\nArgs:\\n query_timeout_seconds (int, optional): Maximum time to wait for the request in seconds.\\n Defaults to DEFAULT_QUERY_TIMEOUT_SECONDS.\\n\\nReturns:\\n str: JSON string conta... | [
{
"role": "user",
"content": "Please delete the container registry authentication with the ID 'auth-12345'."
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_oyisc3yl... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
["{'name': 'parse_stack', 'description': '$20', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['stacks'], 'properties': {'stacks': {'type': 'array', 'items': {'type': 'object', 'required': ['line', 'column', 'sourceMapUrl'], 'properties': {'line': {'type': 'number',... | [
{"role": "system", "content": "You are an intelligent assistant designed to help users with technical tasks. You have access to the following tools:
1. **`parse_stack(stack_trace: str)`**: Use this tool to analyze and understand a provided programming stack trace. This is useful when a user reports an error wit... | INCORRECT | PARAM_VALUE_ERROR | ['The `product_id` in the incorrect parameters is 'MegaPrinter-900', which is subtly different from the user's specified 'MegaPrinter-9000'. This single-digit omission makes it refer to a different or non-existent product ID.'] |
['{\'name\': \'findParks\', \'description\': \'Search for national parks based on state, name, activities, or other criteria\', \'inputSchema\': {\'type\': \'object\', \'$schema\': \'http://json-schema.org/draft-07/schema#\', \'properties\': {\'q\': {\'type\': \'string\', \'description\': \'Search term to filter parks ... | [
{
"role": "user",
"content": "Can you tell me about the visitor centers in Yosemite National Park?"
},
{
"type": "function_call",
"id": "call_ka92hm9u2qx5aqt3vgzea",
"name": "getVisitorCenters",
"arguments": {'parkCode': 'yell'}
}
] | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly asked for information about 'Yosemite National Park', which has the park code 'yose'. The incorrect parameter uses 'yell', which is the park code for Yellowstone National Park, thus requesting information for the wrong park.'] |
["{'type': 'function', 'function': {'name': 'get_biorxiv_metadata', 'description': '', 'parameters': {'properties': {'doi': {'type': 'string', 'title': 'Doi'}}, 'required': ['doi']}}}", "{'type': 'function', 'function': {'name': 'search_biorxiv_advanced', 'description': '', 'parameters': {'properties': {'term': {'anyOf... | [
{"role": "system", "content": "You are an AI assistant designed to interact with Twitter Spaces. You have access to several tools to manage and process Space content.
* `check_space_availability(space_id: str)`: Checks if a Twitter Space is currently live or scheduled.
* `get_transcript(space_id: str)`: Retri... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'analyse_poker_cards', 'description': '$20', 'inputSchema': {'type': 'object', 'title': 'analyse_poker_cardsArguments', 'required': ['my_cards_input', 'community_input', 'opponent_input'], 'properties': {'my_cards_input': {'type': 'string', 'title': 'My Cards Input'}, 'opponent_input': {'type': 'string', 'ti... | [
{"role": "system", "content": "You are a helpful assistant with access to a suite of tools. Use these tools to answer user queries effectively and efficiently. Always prioritize calling the most relevant tool for the user's request.
**Available Tools:**
- `analyse_poker_cards(hand_1: list[str], hand_2: list[st... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly stated the piles are 3, 4, and 5. The incorrect parameters show the last pile as 6, which is a one-number difference from the user's intended value of 5.'] |
["{'name': 'configure_role_alerts', 'description': 'Configure alerts for privileged role assignments (MS.AAD.7.7v1)', 'inputSchema': {'type': 'object', 'required': ['notificationEmails'], 'properties': {'notificationEmails': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Email addresses to notify on rol... | [
{
"role": "user",
"content": "I need to set up an automatic block for any user flagged with a critical risk level. They should be blocked for exactly one week, and make sure security_ops@example.com gets an email notification."
},
{
"completion_message": {
"content... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested a block duration of 'one week', which translates to '7' days. The 'incorrect_tool_call_1_params' specifies '70' days for 'duration_in_days', which is a subtle off-by-one/typo error, making the duration ten times longer than intended.'] |
["{'type': 'function', 'function': {'name': 'tusclasesparticulares_screenshot', 'description': 'Take a screenshot of the current page for debugging', 'parameters': {'properties': {'filename': {'type': 'string', 'description': 'Optional filename for screenshot'}}, 'required': []}}}", '{\'type\': \'function\', \'function... | [
{"role": "system", "content": "You are a helpful assistant with access to several tools. Use them judiciously to fulfill user requests.
- `get_project_by_slug(slug: str)`: Use this to retrieve detailed information about a project when its unique slug (identifier) is provided. This is useful for fetching project-... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'send-email', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['from', 'to'], 'properties': {'cc': {'type': 'string'}, 'to': {'type': 'string'}, 'bcc': {'type': 'string'}, 'from': {'type': 'string', 'format': 'email'}, 'tags': {'anyOf': [{'type': 'string'},... | [
{"role": "system", "content": "You have access to the 'send-email' tool. Use this tool to compose and send emails to specified recipients. When a user asks to send an email, identify the recipient's email address, the subject of the email, and the content for the email body. Call the tool with the 'to', 'subject'... | INCORRECT | PARAM_VALUE_ERROR | ['The 'subject' parameter in the incorrect tool call is 'Project Statues Meeting' (plural 'Statues'), which has a one-letter difference and is a typo. The user query clearly states the subject should be 'Project Status Meeting' (singular 'Status').'] |
["{'type': 'function', 'function': {'name': 'scan_ip_range', 'description': 'Scan a range of IP addresses for MESH systems', 'parameters': {'properties': {'endIp': {'type': 'string'}, 'startIp': {'type': 'string'}}, 'required': ['startIp', 'endIp']}}}", "{'type': 'function', 'function': {'name': 'unlock_entrance', 'des... | [
{
"role": "user",
"content": "Could you please generate a PDF report of the latest network scan results? Please name the file 'network_scan_report_2023-10-27' and send it to admin@example.com."
},
{
"completion_message": {
"content": {
"type": "text... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested the file to be named 'network_scan_report_2023-10-27'. The incorrect parameters specify 'network_scan_results_2023-10-27', substituting 'results' for 'report'. This subtle one-word difference in the 'file_name' parameter deviates from the explicit naming instruction in the user query.'] |
["{'type': 'function', 'function': {'name': 'homeassistant_get_state', 'description': 'Get the state of a Home Assistant entity', 'parameters': {'properties': {'entity_id': {'type': 'string', 'description': 'The entity ID of the Home Assistant entity'}}, 'required': ['entity_id']}}}", "{'type': 'function', 'function': ... | [
{
"role": "user",
"content": "Could you please turn on the living room lamp and set its brightness to 75 percent?"
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_qm... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested the brightness to be set to "75 percent", but the `brightness_pct` parameter in the incorrect call was set to "70", which is a subtle numerical mismatch from the user's instruction.'] |
["{'type': 'function', 'function': {'name': 'exportGameDesign', 'description': 'Export the complete game design as HTML documentation', 'parameters': {'properties': {}, 'required': []}}}", "{'type': 'function', 'function': {'name': 'gamedesignthinking', 'description': '$1f', 'parameters': {'properties': {'shaders': {'t... | [
{"role": "system", "content": "You are a helpful assistant with access to specialized tools for game design.
**Tool List:**
- `exportGameDesign(project_id: str, format: str, version: str, destination: str)`: Use this tool to export a specific game design document.
- `project_id`: The unique identifier for th... | INCORRECT | PARAM_VALUE_ERROR | ['The user requested 'project ID GALAXY_QUEST_42', but the `project_id` parameter is 'GALAXY_QUEST_43' (a one-number difference). Additionally, the user asked for 'version 3.1', but the `version` parameter is 'v3.0' (an off-by-one difference in the minor version number).'] |
["{'name': 'get_trending_gifs', 'description': 'Get currently trending GIFs on Giphy', 'inputSchema': {'type': 'object', 'properties': {'limit': {'type': 'number', 'description': 'Maximum number of objects to return (default: 10, max: 50)'}, 'offset': {'type': 'number', 'description': 'Results offset (default: 0)'}, 'r... | [
{
"role": "user",
"content": "Hey, can you search for some funny dog videos?"
},
{
"id": "call_2b6kmqcbg5egum85gtlwf5fn31pjfr",
"type": "function",
"function": {
"name": "search_gifs",
"arguments": {'query': 'funny dod videos'}
}... | INCORRECT | PARAM_VALUE_ERROR | ['The 'query' parameter contains a subtle typo, 'dod' instead of 'dog', which is a one-letter difference that would lead to incorrect search results not matching the user's request for 'funny dog videos'.'] |
["{'name': 'transfer_funds', 'description': 'Transfer funds to another Stellar wallet', 'inputSchema': {'type': 'object', 'required': ['secretKey', 'destinationAddress', 'amount'], 'properties': {'asset': {'type': 'string', 'description': 'Asset to transfer (defaults to XLM)'}, 'amount': {'type': 'string', 'description... | [
{
"role": "user",
"content": "Could you list all the pages in the 'DOCS' Confluence space for me?"
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_da4kum9pjflz8ncrwh... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
['{\'type\': \'function\', \'function\': {\'name\': \'create_entity\', \'description\': \'Create Fibery entity with specified fields.\\nExamples (note, that these databases are non-existent, use databases only from user\\\'s schema!):\\nQuery: Create a feature\\nTool use:\\n{\\n "database": "Product Management/Featu... | [
{"role": "system", "content": "You are an intelligent assistant designed to interact with a database system. You have access to the following tools:
- "create_entity(table_name: str, data: dict)": Use this tool to add new records to a specified database table. Provide the table name and a dictionary of the data t... | INCORRECT | PARAM_VALUE_ERROR | ['The user specifically asked for products that cost 'less than $500'. Given that the `price_max` filter is typically inclusive (as stated in the system prompt), the correct parameter should be `price_max: 499` to ensure products priced at exactly $500 are excluded. The incorrect parameters use `price_max: 500`, which ... |
['{\'name\': \'issue_get_comments\', \'description\': \'Get comments of a Yandex Tracker issue by its id\', \'inputSchema\': {\'type\': \'object\', \'title\': \'issue_get_commentsArguments\', \'required\': [\'issue_id\'], \'properties\': {\'issue_id\': {\'type\': \'string\', \'title\': \'Issue Id\', \'description\': "I... | [
{"role": "system", "content": "You are an AI assistant designed to help developers debug GitHub Actions workflows. You have access to the following tools:
- "get-failed-workflow-runs(owner: string, repo: string)": Use this to list recent failed workflow runs for a given repository. This is useful when a user ask... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
['{\'name\': \'write-research-file\', \'description\': "Write research content to a file. This tool allows you to save research findings, documentation, or any text content to a specified file path.\\n\\n SECURITY: File writing is controlled by environment variables:\\n - FILE_WRIT... | [
{
"role": "user",
"content": "Can you find all reports on renewable energy authored by Dr. Anya Sharma published in 2023?"
},
{
"type": "function_call",
"id": "call_dx060ogi",
"name": "metadata_search",
"arguments": {'document_type': 'report', 'topic': ... | INCORRECT | PARAM_VALUE_ERROR | ['The incorrect parameters specify the author as 'Dr. Ana Sharma', which is a subtle one-letter difference (Ana vs. Anya) from the requested author 'Dr. Anya Sharma' in the user query. This small typo would lead to an incorrect search result or no results for the intended author.'] |
["{'name': 'extract_key_moments', 'description': 'Extracts key moments (timestamps and descriptions) from a given YouTube video.', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['youtube_url'], 'properties': {'youtube_url': {'type': 'string', 'format': 'uri'}, 'numb... | [
{"role": "system", "content": "You have access to a financial calculator tool named "emi". Use this tool to calculate the Equated Monthly Installment (EMI) for a loan. When using the "emi" tool, you must provide the "principal_amount" (the total loan amount), the "interest_rate" (the annual interest rate as a per... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'type': 'function', 'function': {'name': 'reasoning', 'description': 'Reasoning operations', 'parameters': {'properties': {}, 'required': []}}}", "{'name': 'visualization', 'annotations': {'_def': {'options': [{'_def': {'catchall': {'_def': {'typeName': 'ZodNever'}, '~standard': {'vendor': 'zod', 'version': 1}}, 'ty... | [
{
"role": "user",
"content": "I need to perform a reasoning operation to analyze this data."
},
{
"type": "function_call",
"id": "call_ppq25h0pjtup7xm3dxtxrxms",
"name": "reasoning",
"arguments": {'enable_deep_thought': 'yes_please'}
}
] | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
['{\'name\': \'get_flight_data\', \'description\': \'Get real-time data for a specific flight by flight number\', \'inputSchema\': {\'type\': \'object\', \'oneOf\': [{\'required\': [\'flight_iata\']}, {\'required\': [\'flight_icao\']}], \'properties\': {\'flight_iata\': {\'type\': \'string\', \'description\': "IATA fli... | [
{"role": "system", "content": "You are a helpful assistant with access to several tools. Use them to assist the user by understanding their intent and calling the appropriate function with the correct parameters. Do not make up information or use tools for unrelated requests.
Available tools:
- `add_to_reminders... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'type': 'function', 'function': {'name': 'search_meal', 'description': '\\n Search for a meal by name and return cooking instructions.\\n ', 'parameters': {'properties': {'meal_name': {'type': 'string', 'title': 'Meal Name'}}, 'required': ['meal_name']}}}", "{'type': 'function', 'function': {'name': 'flight-de... | [
{"role": "system", "content": "You are a helpful assistant with access to several specialized tools. Use these tools wisely to fulfill user requests.
Available tools:
- 'search_meal(query: str)': Use this to find recipes or meal ideas based on ingredients, cuisine, or meal type.
- 'flight-delay-compensation-chec... | INCORRECT | PARAM_VALUE_ERROR | ['The incorrect parameter 'algorthim' contains a subtle spelling mistake (swapped 'o' and 'r' after 'g') compared to the correct term 'algorithm' from the user's query. This minor typo would likely result in an empty or incorrect definition being returned.'] |
["{'type': 'function', 'function': {'name': 'cal', 'description': 'Use the expr-eval library to evaluate the input mathematical expression and return the result.\\n\\nConstant \\tDescription\\nE \\tThe value of Math.E from your JavaScript runtime\\nPI \\tThe value of Math.PI from your JavaScript runtime\\ntrue \\tLogic... | [
{
"role": "user",
"content": "What does the ICD-10-CM code I10 refer to?"
},
{
"id": "call_6owbkipvwj6fdv4efkrk",
"type": "function",
"function": {
"name": "lookup_icd_code",
"arguments": {'code': 'J10', 'version': 'ICD-10-CM'}
}... | INCORRECT | PARAM_VALUE_ERROR | ['The 'code' parameter in the incorrect tool call ('J10') has a one-letter difference from the user's query ('I10'). The correct code should be 'I10' as explicitly stated by the user.'] |
['{\'name\': \'k8s_get_documentation\', \'description\': \'$21\', \'inputSchema\': {\'type\': \'object\', \'$schema\': \'http://json-schema.org/draft-07/schema#\', \'required\': [\'query\'], \'properties\': {\'query\': {\'type\': \'string\', \'description\': \'A natural language question (e.g., "How do I define a Deplo... | [
{
"role": "user",
"content": "I need to know how to perform a left join using the pandas library, specifically for version 2.1.0. Can you find the documentation for that?"
},
{
"id": "call_wlexl0iwjsntxfrjff69wqkhgsr97",
"type": "function",
"function": {
... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
['{\'name\': \'jira_search\', \'description\': \'Search Jira issues with JQL\', \'inputSchema\': {\'type\': \'object\', \'$schema\': \'http://json-schema.org/draft-07/schema#\', \'required\': [\'jql\'], \'properties\': {\'jql\': {\'type\': \'string\', \'description\': \'JQL query (e.g., "project = PROJ AND status = Ope... | [
{
"role": "user",
"content": "Can you fetch the details for the Jira issue PROJ-123?"
},
{
"type": "function_call",
"id": "call_yvmq7buvcv3",
"name": "jira_get_issue",
"arguments": {'issueKey': 123}
}
] | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
["{'type': 'function', 'function': {'name': 'fetch_txt', 'description': 'Fetch a website, return the content as plain text (no HTML)', 'parameters': {'properties': {'url': {'type': 'string', 'description': 'URL of the website to fetch'}, 'headers': {'type': 'object', 'description': 'Optional headers to include in the r... | [
{"role": "system", "content": "You have access to a set of tools for fetching content from URLs. Choose the appropriate tool based on the user's request for the content type:
- `fetch_txt(url: str)`: Use this to retrieve plain text content from a given URL.
- `fetch_markdown(url: str)`: Use this to retrieve conte... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested product ID '123', but the incorrect parameters specify product ID '124'. This is an off-by-one ID mismatch from the user's specific query.'] |
['{\'type\': \'function\', \'function\': {\'name\': \'image_analysis\', \'description\': \'$20\', \'parameters\': {\'properties\': {\'image\': {\'type\': \'string\', \'title\': \'Image\'}, \'model\': {\'anyOf\': [{\'type\': \'string\'}, {\'type\': \'null\'}], \'title\': \'Model\', \'default\': None}, \'query\': {\'type... | [
{"role": "system", "content": "You are an AI assistant designed to help users manage their credentials. You have access to the following tools:
1. `get_password(service_name: str, username: str)`: Use this tool to retrieve the password for a specified service and username. Call this tool when the user explicitly... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
["{'type': 'function', 'function': {'name': 'list_resources', 'description': 'List resources from a specific category', 'parameters': {'properties': {'category': {'enum': ['OBJECTS', 'POLICIES', 'NETWORK', 'DEVICES'], 'type': 'string', 'description': 'Resource category to list'}, 'resource_type': {'type': 'string', 'de... | [
{
"role": "user",
"content": "Could you please show me the current settings for the primary payment gateway in the production environment?"
},
{
"completion_message": {
"content": {
"type": "text",
"text": ""
},
... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly asked for the 'primary payment gateway', which maps to 'payment_gateway_config_001'. The incorrect parameter uses 'payment_gateway_config_01', which is a subtle one-digit difference (missing a '0') that might refer to a different or non-existent configuration node, leading to an incorrect tool cal... |
['{\'name\': \'web_search\', \'inputSchema\': {\'type\': \'object\', \'$schema\': \'http://json-schema.org/draft-07/schema#\', \'required\': [\'query\'], \'properties\': {\'query\': {\'type\': \'string\', \'description\': \'The search query to look up\'}, \'region\': {\'type\': \'string\', \'description\': "Region code... | [
{"role": "system", "content": "You are an AI assistant with access to a `web_search` tool. Use `web_search` to find information on the internet when a user's request requires current events, factual data, definitions, statistics, or general knowledge that is not pre-programmed into your training data. Always form... | INCORRECT | PARAM_VALUE_ERROR | ['The 'query' parameter contains a subtle misspelling of 'weather' as 'weatther,' which is a one-letter difference and would likely lead to less accurate or no relevant search results.'] |
["{'name': 'get_pubmed_count', 'description': '\\n Get the number of PubMed results for multiple search terms.\\n \\n This function queries PubMed and returns the count of results for each provided search term.\\n Useful for comparing the prevalence of different medical terms or concepts in the literature.\... | [
{"role": "system", "content": "You are a specialized research assistant with access to powerful tools for biomedical literature search. Your goal is to accurately interpret user requests and use the most appropriate tool(s) to retrieve relevant information.
Available tools:
- `get_pubmed_count(query: str)`: Use... | INCORRECT | PARAM_VALUE_ERROR | ['The 'intervention' parameter in the incorrect tool call uses 'cognitive behavior therapy' (missing 'al') instead of 'cognitive behavioral therapy', which was explicitly mentioned in the user's query.'] |
["{'name': 'post_text_message', 'description': 'Post a text message to a Google Chat space.', 'inputSchema': {'type': 'object', 'required': ['text'], 'properties': {'text': {'type': 'string', 'description': 'The text content of the message to be posted.'}}}}", "{'type': 'function', 'function': {'name': 'send_summary', ... | [
{"role": "system", "content": "You are an expert SEO assistant specializing in Google Search Console. Your primary role is to help users manage their website's presence on Google Search by leveraging a comprehensive set of tools. You can provide insights into site performance, indexing status, sitemaps, and searc... | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
["{'type': 'function', 'function': {'name': 'create_prediction', 'description': 'Generate an prediction from a text prompt using Flux Schnell model', 'parameters': {'properties': {'seed': {'type': 'integer', 'description': 'Random seed. Set for reproducible generation'}, 'prompt': {'type': 'string', 'minLength': 1, 'de... | [
{
"role": "user",
"content": "Could you create a high-definition, vivid image for me? I'm thinking of a futuristic cityscape at sunset, complete with flying cars and dazzling neon signs. Use the Dall-E 3 model, and make sure it's in a widescreen format."
},
{
"role": "assistan... | INCORRECT | PARAM_VALUE_ERROR | ['The user requested a 'widescreen format' for the image. While '1792x1020' is technically widescreen, '1792x1024' is the standard and explicitly supported widescreen resolution for the 'dall-e-3' model. The provided '1792x1020' is an off-by-four value, making it an invalid or non-standard resolution for the intended m... |
["{'type': 'function', 'function': {'name': 'get-notes', 'description': 'Get notes', 'parameters': {'properties': {'end_date': {'type': 'string'}, 'start_date': {'type': 'string'}}, 'required': []}}}", "{'name': 'add-habit-log', 'description': 'Add a habit log', 'inputSchema': {'type': 'object', '$schema': 'http://json... | [
{"role": "system", "content": "You are an AI assistant with access to the following tools:
- visit_page(url: str): Use this tool to navigate to a specific web page. Provide the full URL as a string, including the protocol (e.g., 'https://').
- search_google(query: str): Use this tool to perform a Google search fo... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'decrypt_message', 'description': 'Decrypt an encrypted message received from a specific Hive account', 'inputSchema': {'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['encrypted_message', 'sender'], 'properties': {'sender': {'type': 'string', 'description': 'Hive usernam... | [
{"role": "system", "content": "You are a helpful assistant with access to tools. You have the following tools available:
- "get_forecast(location: str, unit: str = 'fahrenheit')": Get the current weather forecast for a specified location. The `unit` parameter can be 'celsius' or 'fahrenheit'.
- "get_alerts(locati... | INCORRECT | PARAM_VALUE_ERROR | ['wrong data type'] |
["{'name': 'add', 'description': 'Add two numbers\\nArgs:\\n a: The first number\\n b: The second number\\nReturns:\\n The sum of the two numbers\\n', 'inputSchema': {'type': 'object', 'title': 'addArguments', 'required': ['a', 'b'], 'properties': {'a': {'type': 'integer', 'title': 'A'}, 'b': {'type': 'integer... | [
{
"role": "user",
"content": "In project 'web-app' with ID 12345, please update the file located at `src/config.js` on the `dev` branch. The new content should be `export const API_ENDPOINT = 'https://api.prod.com';`. Use the commit message 'Update API endpoint to production'."
},
{
... | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
['{\'name\': \'supadata_map\', \'description\': \'Crawl a whole website and get all URLs on it using Supadata\\\'s mapping API.\\n\\n**Purpose:** Extract all links found on a website for content discovery and sitemap creation.\\n**Best for:** Website content discovery, SEO analysis, content aggregation, automated web s... | [
{
"role": "user",
"content": "Can you start a crawl on the blog section of example.com? I only need to go two levels deep, and please make sure it stays strictly within the example.com domain."
},
{
"completion_message": {
"content": {
"type": "text... | INCORRECT | PARAM_VALUE_ERROR | ['The user specifically requested to crawl the 'blog section', but the provided 'start_url' in the incorrect parameters points to '/vlog' instead of '/blog', which is a single letter difference leading to an incorrect target path for the crawl.'] |
["{'type': 'function', 'function': {'name': 'busca_cep', 'description': 'Busca o cep de um cliente', 'parameters': {'properties': {'cep': {'type': 'string', 'description': 'Número do CEP'}}, 'required': ['cep']}}}", "{'name': 'player_seek_to', 'description': 'Seeks to a specific position in time for a MultiViewer view... | [
{"role": "system", "content": "You are an AI assistant designed to manage alerting and silencing. You have access to the following tools:
- `delete-silence(silence_id: str)`: Deletes an existing alert silence.
- `get-alert-details(alert_id: str)`: Retrieves detailed information for a specific alert.
- `create-sil... | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
['{\'name\': \'API-coins-list\', \'description\': \'Coins List (ID Map)\', \'inputSchema\': {\'type\': \'object\', \'$defs\': {\'Ping\': {\'type\': \'object\', \'properties\': {\'gecko_says\': {\'type\': \'string\'}}, \'additionalProperties\': True}, \'Global\': {\'type\': \'object\', \'properties\': {\'data\': {\'type... | [
{"role": "system", "content": "You are an AI assistant capable of interacting with a cryptocurrency data API. You have access to the following tools:
* `API-coins-list()`: Use this tool to retrieve a comprehensive list of all supported cryptocurrencies, including their unique IDs, symbols, and full names. Call... | INCORRECT | PARAM_VALUE_ERROR | ['The incorrect parameter 'bitcoiin' contains an extra 'i', making it a misspelling of 'bitcoin' which was clearly intended by the user's query. This subtle typo would likely lead to an incorrect search result or no result at all.'] |
['{\'name\': \'get_deep_directory_tree\', \'description\': \'Get deep directory tree\', \'inputSchema\': {\'type\': \'object\', \'$schema\': \'http://json-schema.org/draft-07/schema#\', \'required\': [\'path\'], \'properties\': {\'path\': {\'type\': \'string\', \'description\': \'Path to get the directory tree (prefera... | [
{
"role": "user",
"content": "Can you help me locate Alice Smith? I believe she works in the Marketing department."
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"type": "tool_use",
"id": "call_m... | INCORRECT | PARAM_VALUE_ERROR | ['The 'department' parameter in the incorrect tool call is 'Marketng', which is missing the 'i' from 'Marketing' as specified in the user's query. This is a subtle one-letter typo.'] |
['{\'type\': \'function\', \'function\': {\'name\': \'start_project\', \'description\': "Retrieves the project details and the prompt for the very first task of a specified project using the project\'s unique slug (e.g., \'CRD\'). This is useful for initiating work on a new project or understanding its starting point."... | [
{
"role": "user",
"content": "Get me all the information available for task TSK-101."
},
{
"type": "function_call",
"id": "call_sj815zowkq",
"name": "get_task",
"arguments": {'number': 101}
}
] | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
["{'type': 'function', 'function': {'name': 'search_airbnb_listings', 'description': '$1f', 'parameters': {'properties': {'pets': {'type': 'integer', 'title': 'Pets', 'default': 0}, 'place': {'type': 'string', 'title': 'Place'}, 'adults': {'type': 'integer', 'title': 'Adults', 'default': 1}, 'infants': {'type': 'intege... | [
{"role": "system", "content": "You are an intelligent assistant designed to help manage and monitor security operations. You have access to the following tools:
1. `list_security_rules(status: str, category: str, owner: str, environment: str = 'production', limit: int = 100)`: Use this tool to retrieve a list of... | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
["{'name': 'remote_macos_mouse_move', 'description': 'Move the mouse cursor to specified coordinates on a remote MacOs machine, with automatic coordinate scaling. Uses environment variables for connection details.', 'inputSchema': {'type': 'object', 'required': ['x', 'y'], 'properties': {'x': {'type': 'integer', 'descr... | [
{
"role": "user",
"content": "On the remote Mac, please move the mouse cursor to coordinates X 500 and Y 350, assuming a source screen resolution of 1920x1080."
},
{
"role": "assistant",
"stop_reason": "tool_use",
"content": [
{
"typ... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
['{\'name\': \'get_realtime_data\', \'description\': "Get real-time stock market data. \'eastmoney_direct\' support all A,B,H shares", \'inputSchema\': {\'type\': \'object\', \'properties\': {\'source\': {\'enum\': [\'xueqiu\', \'eastmoney\', \'eastmoney_direct\'], \'type\': \'string\', \'title\': \'Source\', \'default... | [
{
"role": "user",
"content": "What's the current stock price of Apple?"
},
{
"id": "call_6z2qbpcs7q9d5i",
"type": "function",
"function": {
"name": "get_realtime_data",
"arguments": {'symbol': 'APPL'}
}
}
] | INCORRECT | PARAM_VALUE_ERROR | ['The user query asks for 'Apple's' stock price. The correct ticker symbol for Apple is 'AAPL'. The incorrect parameter uses 'APPL', which is a common typo for Apple but not its actual stock ticker. This would lead to an invalid or no result if 'APPL' does not represent a valid stock.'] |
['{\'type\': \'function\', \'function\': {\'name\': \'mmr_search\', \'description\': \'\\nSearch a PGVector collection using MMR (Maximal Marginal Relevance) search.\\n\\nThe connected vector database contains comprehensive restaurant information \\nfor Seoul, South Korea, including restaurant names, addresses, cuisine... | [
{"role": "system", "content": "You are a helpful assistant with access to specialized tools.
1. `mmr_search(query: str, filters: dict = None, top_k: int = 5)`: Use this tool to perform a general search across various internal documents and knowledge bases. Provide a concise `query` string. Optional `filters` can ... | INCORRECT | PARAM_VALUE_ERROR | ['The 'filters' dictionary contains a typo: 'companie' instead of the correct key 'company'. This subtle one-letter difference would prevent the filter from correctly identifying results for 'Acme Corp' as specified in the user query.'] |
["{'name': 'show_create_table', 'description': 'Show the CREATE TABLE statement for a specific table', 'inputSchema': {'type': 'object', 'title': 'show_create_tableArguments', 'required': ['table'], 'properties': {'table': {'type': 'string', 'title': 'Table'}, 'schema': {'anyOf': [{'type': 'string'}, {'type': 'null'}],... | [
{
"role": "user",
"content": "I need to see the DDL for the 'orders' table, which is located in the 'production' schema within the 'hive' catalog."
},
{
"type": "function_call",
"id": "call_e8d3jehw066gv0xrbp1b70fc854",
"name": "show_create_table",
"arg... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'fetch_plaintext', 'description': 'Fetch a website and return the content as plain text with HTML tags removed. Best practices: 1) Always set startCursor=0 for initial requests, and use the fetchedBytes value from previous response for subsequent requests to ensure content continuity. 2) Set contentSizeLimit... | [
{"role": "system", "content": "You have access to the following tools to retrieve information: 'fetch_plaintext', 'fetch_json', 'fetch_txt', 'fetch_html'.
- Use 'fetch_plaintext(url: string)' when you need to retrieve generic unformatted text content from a specified URL.
- Use 'fetch_json(url: string)' when you... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested the log for 'November 15th, 2023', but the incorrect parameter points to '2023-11-16', which is an off-by-one day mismatch.'] |
["{'type': 'function', 'function': {'name': 'create_prediction', 'description': 'Generate an prediction from a text prompt using Flux Schnell model', 'parameters': {'properties': {'seed': {'type': 'integer', 'description': 'Random seed. Set for reproducible generation'}, 'prompt': {'type': 'string', 'minLength': 1, 'de... | [
{
"role": "user",
"content": "Can you generate three high-resolution images of a futuristic city at sunset, complete with flying cars and neon lights?"
},
{
"completion_message": {
"content": {
"type": "text",
"text": ""
... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested 'three' images, but the incorrect tool call specifies 'count: 4', which is an off-by-one error. Additionally, the user's prompt mentioned 'futuristic city', but the incorrect tool call contains a typo, 'futuristic cty', which is a one-letter difference.'] |
["{'name': 'youtube_transcription', 'description': 'Extracts transcription from a Youtube video provided by the user', 'inputSchema': {'type': 'object', 'required': ['link'], 'properties': {'link': {'type': 'string', 'description': 'Youtube video link'}}}}"] | [
{
"role": "user",
"content": "Could you please get me the full transcript for the YouTube video that has the ID 'k2M_y8sFv3I'? It's an introduction to Python."
},
{
"type": "function_call",
"id": "call_zcgx7vt",
"name": "youtube_transcription",
"argumen... | INCORRECT | PARAM_VALUE_ERROR | ['The 'video_id' parameter in the incorrect call uses a lowercase 'l' instead of an uppercase 'I' at the end ('k2M_y8sFv3l' vs 'k2M_y8sFv3I'), which is a subtle but crucial difference from the ID specified in the user's query.'] |
["{'name': 'todoist_rename_shared_labels', 'description': 'Rename one or more shared labels in Todoist', 'inputSchema': {'type': 'object', 'anyOf': [{'required': ['labels']}, {'required': ['name', 'new_name']}], 'properties': {'name': {'type': 'string', 'description': 'The name of the existing label to rename'}, 'label... | [
{
"role": "user",
"content": "Show me all my overdue tasks with priority 1."
},
{
"id": "call_4yrap",
"type": "function",
"function": {
"name": "todoist_get_tasks",
"arguments": {'filter': True, 'priority': 'highest'}
}
}
] | INCORRECT | PARAM_VALUE_ERROR | ['incorrect data type'] |
["{'name': 'text-translation', 'description': 'Translate text quickly and easily with JigsawStack, we support over 180+ language pairs.', 'inputSchema': {'type': 'object', 'required': ['text', 'target_language'], 'properties': {'text': {'type': 'string', 'description': 'Text in current langauge.'}, 'target_language': {... | [
{"role": "system", "content": "You are an AI assistant with access to two translation tools: 'text-translation' and 'image-translation'.
1. Use 'text-translation' when the user provides plain text that needs to be translated. This tool requires 'text_to_translate' (string), 'source_language' (ISO 639-1 code, e.g... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly mentioned a 'Spanish restaurant menu', which clearly indicates 'es' (Spanish) as the source language. The incorrect parameters use 'et' (Estonian) for the source_language, which is a one-letter difference that makes it subtly but incorrectly matched to the user's request.'] |
["{'name': 'get_hero_stats', 'description': 'Get statistics for heroes.\\n\\n Args:\\n hero_id: Optional hero ID to get stats for a specific hero\\n\\n Returns:\\n Hero statistics including win rates by skill bracket\\n ', 'inputSchema': {'type': 'object', 'title': 'get_hero_statsArguments', 'pro... | [
{"role": "system", "content": "You are an AI assistant with access to several tools. Your goal is to use these tools accurately and efficiently to fulfill user requests.
- **greet(name: str)**: Use this tool to greet a user or a specified person by their name. Call it when the user explicitly asks for a greetin... | INCORRECT | PARAM_VALUE_ERROR | ['contextually inappropriate'] |
["{'name': 'execute_query', 'description': 'Executes a Kusto Query Language (KQL) query against the configured Azure Data Explorer database and returns the results as a list of dictionaries.', 'inputSchema': {'type': 'object', 'title': 'execute_queryArguments', 'required': ['query'], 'properties': {'query': {'type': 's... | [
{
"role": "user",
"content": "I want to book a round trip flight from JFK to LAX for two people, departing on Christmas Day this year and returning on January 5th of next year. We're looking for economy class tickets."
},
{
"id": "call_m1l8rilbxpobpbaet75s7vzchqnc6",
"... | INCORRECT | PARAM_VALUE_ERROR | ['The user explicitly requested to depart on 'Christmas Day this year', which translates to '2024-12-25'. The incorrect parameter '2024-12-26' is off by one day and does not match the user's specified departure date.'] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.