mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-10-16 14:57:20 +00:00
e9b4278a51
This PR updates the Conversation item related types and improves a couple critical parts of the implemenation: - it creates a streaming output item for the final assistant message output by the model. until now we only added content parts and included that message in the final response. - rewrites the conversation update code completely to account for items other than messages (tool calls, outputs, etc.) ## Test Plan Used the test script from https://github.com/llamastack/llama-stack-client-python/pull/281 for this ``` TEST_API_BASE_URL=http://localhost:8321/v1 \ pytest tests/integration/test_agent_turn_step_events.py::test_client_side_function_tool -xvs ```
320 lines
10 KiB
JSON
320 lines
10 KiB
JSON
{
|
|
"test_id": "tests/integration/responses/test_tool_responses.py::test_response_streaming_multi_turn_tool_execution[client_with_models-txt=openai/gpt-4o-user_permissions_workflow]",
|
|
"request": {
|
|
"method": "POST",
|
|
"url": "https://api.openai.com/v1/v1/chat/completions",
|
|
"headers": {},
|
|
"body": {
|
|
"model": "gpt-4o",
|
|
"messages": [
|
|
{
|
|
"role": "user",
|
|
"content": "Help me with this security check: First, get the user ID for 'charlie', then get the permissions for that user ID, and finally check if that user can access 'secret_file.txt'. Stream your progress as you work through each step. Return only one tool call per step. Summarize the final result with a single 'yes' or 'no' response."
|
|
},
|
|
{
|
|
"role": "assistant",
|
|
"content": "",
|
|
"tool_calls": [
|
|
{
|
|
"index": 0,
|
|
"id": "call_mEjLRxcZhRdCdW0Ibeu6Pkju",
|
|
"type": "function",
|
|
"function": {
|
|
"name": "get_user_id",
|
|
"arguments": "{\"username\":\"charlie\"}"
|
|
}
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"role": "tool",
|
|
"tool_call_id": "call_mEjLRxcZhRdCdW0Ibeu6Pkju",
|
|
"content": [
|
|
{
|
|
"type": "text",
|
|
"text": "user_11111"
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"role": "assistant",
|
|
"content": "",
|
|
"tool_calls": [
|
|
{
|
|
"index": 0,
|
|
"id": "call_h85dbcBdBPF5ti1xrNcnnaEv",
|
|
"type": "function",
|
|
"function": {
|
|
"name": "get_user_permissions",
|
|
"arguments": "{\"user_id\":\"user_11111\"}"
|
|
}
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"role": "tool",
|
|
"tool_call_id": "call_h85dbcBdBPF5ti1xrNcnnaEv",
|
|
"content": [
|
|
{
|
|
"type": "text",
|
|
"text": "admin"
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"role": "assistant",
|
|
"content": "",
|
|
"tool_calls": [
|
|
{
|
|
"index": 0,
|
|
"id": "call_3OfzDbR92CZR2kvccYE65oDr",
|
|
"type": "function",
|
|
"function": {
|
|
"name": "check_file_access",
|
|
"arguments": "{\"user_id\":\"user_11111\",\"filename\":\"secret_file.txt\"}"
|
|
}
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"role": "tool",
|
|
"tool_call_id": "call_3OfzDbR92CZR2kvccYE65oDr",
|
|
"content": [
|
|
{
|
|
"type": "text",
|
|
"text": "no"
|
|
}
|
|
]
|
|
}
|
|
],
|
|
"stream": true,
|
|
"stream_options": {
|
|
"include_usage": true
|
|
},
|
|
"tools": [
|
|
{
|
|
"type": "function",
|
|
"function": {
|
|
"name": "get_user_id",
|
|
"description": "\n Get the user ID for a given username. This ID is needed for other operations.\n\n :param username: The username to look up\n :return: The user ID for the username\n ",
|
|
"parameters": {
|
|
"properties": {
|
|
"username": {
|
|
"title": "Username",
|
|
"type": "string"
|
|
}
|
|
},
|
|
"required": [
|
|
"username"
|
|
],
|
|
"title": "get_user_idArguments",
|
|
"type": "object"
|
|
}
|
|
}
|
|
},
|
|
{
|
|
"type": "function",
|
|
"function": {
|
|
"name": "get_user_permissions",
|
|
"description": "\n Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n :param user_id: The user ID to check permissions for\n :return: The permissions for the user\n ",
|
|
"parameters": {
|
|
"properties": {
|
|
"user_id": {
|
|
"title": "User Id",
|
|
"type": "string"
|
|
}
|
|
},
|
|
"required": [
|
|
"user_id"
|
|
],
|
|
"title": "get_user_permissionsArguments",
|
|
"type": "object"
|
|
}
|
|
}
|
|
},
|
|
{
|
|
"type": "function",
|
|
"function": {
|
|
"name": "check_file_access",
|
|
"description": "\n Check if a user can access a specific file. Requires a valid user ID.\n\n :param user_id: The user ID to check access for\n :param filename: The filename to check access to\n :return: Whether the user can access the file (yes/no)\n ",
|
|
"parameters": {
|
|
"properties": {
|
|
"user_id": {
|
|
"title": "User Id",
|
|
"type": "string"
|
|
},
|
|
"filename": {
|
|
"title": "Filename",
|
|
"type": "string"
|
|
}
|
|
},
|
|
"required": [
|
|
"user_id",
|
|
"filename"
|
|
],
|
|
"title": "check_file_accessArguments",
|
|
"type": "object"
|
|
}
|
|
}
|
|
},
|
|
{
|
|
"type": "function",
|
|
"function": {
|
|
"name": "get_experiment_id",
|
|
"description": "\n Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n :param experiment_name: The name of the experiment\n :return: The experiment ID\n ",
|
|
"parameters": {
|
|
"properties": {
|
|
"experiment_name": {
|
|
"title": "Experiment Name",
|
|
"type": "string"
|
|
}
|
|
},
|
|
"required": [
|
|
"experiment_name"
|
|
],
|
|
"title": "get_experiment_idArguments",
|
|
"type": "object"
|
|
}
|
|
}
|
|
},
|
|
{
|
|
"type": "function",
|
|
"function": {
|
|
"name": "get_experiment_results",
|
|
"description": "\n Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n :param experiment_id: The experiment ID to get results for\n :return: The experiment results\n ",
|
|
"parameters": {
|
|
"properties": {
|
|
"experiment_id": {
|
|
"title": "Experiment Id",
|
|
"type": "string"
|
|
}
|
|
},
|
|
"required": [
|
|
"experiment_id"
|
|
],
|
|
"title": "get_experiment_resultsArguments",
|
|
"type": "object"
|
|
}
|
|
}
|
|
}
|
|
]
|
|
},
|
|
"endpoint": "/v1/chat/completions",
|
|
"model": "gpt-4o"
|
|
},
|
|
"response": {
|
|
"body": [
|
|
{
|
|
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
|
|
"__data__": {
|
|
"id": "rec-a9c49a7e865b",
|
|
"choices": [
|
|
{
|
|
"delta": {
|
|
"content": "",
|
|
"function_call": null,
|
|
"refusal": null,
|
|
"role": "assistant",
|
|
"tool_calls": null
|
|
},
|
|
"finish_reason": null,
|
|
"index": 0,
|
|
"logprobs": null
|
|
}
|
|
],
|
|
"created": 0,
|
|
"model": "gpt-4o-2024-08-06",
|
|
"object": "chat.completion.chunk",
|
|
"service_tier": "default",
|
|
"system_fingerprint": "fp_eb3c3cb84d",
|
|
"usage": null,
|
|
"obfuscation": "KQwtNGSrFd9uWd"
|
|
}
|
|
},
|
|
{
|
|
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
|
|
"__data__": {
|
|
"id": "rec-a9c49a7e865b",
|
|
"choices": [
|
|
{
|
|
"delta": {
|
|
"content": "No",
|
|
"function_call": null,
|
|
"refusal": null,
|
|
"role": null,
|
|
"tool_calls": null
|
|
},
|
|
"finish_reason": null,
|
|
"index": 0,
|
|
"logprobs": null
|
|
}
|
|
],
|
|
"created": 0,
|
|
"model": "gpt-4o-2024-08-06",
|
|
"object": "chat.completion.chunk",
|
|
"service_tier": "default",
|
|
"system_fingerprint": "fp_eb3c3cb84d",
|
|
"usage": null,
|
|
"obfuscation": "OcpUScMebTHPiu"
|
|
}
|
|
},
|
|
{
|
|
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
|
|
"__data__": {
|
|
"id": "rec-a9c49a7e865b",
|
|
"choices": [
|
|
{
|
|
"delta": {
|
|
"content": null,
|
|
"function_call": null,
|
|
"refusal": null,
|
|
"role": null,
|
|
"tool_calls": null
|
|
},
|
|
"finish_reason": "stop",
|
|
"index": 0,
|
|
"logprobs": null
|
|
}
|
|
],
|
|
"created": 0,
|
|
"model": "gpt-4o-2024-08-06",
|
|
"object": "chat.completion.chunk",
|
|
"service_tier": "default",
|
|
"system_fingerprint": "fp_eb3c3cb84d",
|
|
"usage": null,
|
|
"obfuscation": "Zp1Ngwd9co"
|
|
}
|
|
},
|
|
{
|
|
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
|
|
"__data__": {
|
|
"id": "rec-a9c49a7e865b",
|
|
"choices": [],
|
|
"created": 0,
|
|
"model": "gpt-4o-2024-08-06",
|
|
"object": "chat.completion.chunk",
|
|
"service_tier": "default",
|
|
"system_fingerprint": "fp_eb3c3cb84d",
|
|
"usage": {
|
|
"completion_tokens": 2,
|
|
"prompt_tokens": 542,
|
|
"total_tokens": 544,
|
|
"completion_tokens_details": {
|
|
"accepted_prediction_tokens": 0,
|
|
"audio_tokens": 0,
|
|
"reasoning_tokens": 0,
|
|
"rejected_prediction_tokens": 0
|
|
},
|
|
"prompt_tokens_details": {
|
|
"audio_tokens": 0,
|
|
"cached_tokens": 0
|
|
}
|
|
},
|
|
"obfuscation": "2ukiTHSsZZx8dP"
|
|
}
|
|
}
|
|
],
|
|
"is_streaming": true
|
|
},
|
|
"id_normalization_mapping": {}
|
|
}
|