phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-31 05:23:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: RFC: tools API rework

Browse source 
# What does this PR do?
This PR proposes updates to the tools API in Inference and Agent.

Goals:
1. Agent's tool specification should be consistent with Inference's tool spec, but with add-ons.
2. Formal types should be defined for built in tools. Currently Agent tools args are untyped, e.g. how does one know that `builtin::rag_tool` takes a `vector_db_ids` param or even how to know 'builtin::rag_tool' is even available (in code, outside of docs)?

Inference:
1. BuiltinTool is to be removed and replaced by a formal `type` parameter.
2. 'brave_search' is replaced by 'web_search' to be more generic. It will still be translated back to brave_search when the prompt is constructed to be consistent with model training.
3. I'm not sure what `photogen` is. Maybe it can be removed?

Agent:
1. Uses the same format as in Inference for builtin tools.
2. New tools types are added, i.e. knowledge_sesarch (currently rag_tool), and MCP tool.
3. Toolgroup as a concept will be removed since it's really only used for MCP.
4. Instead MCPTool is its own type and available tools provided by the server will be expanded by default. Users can specify a subset of tool names if desired.

Example snippet:
```

agent = Agent(
    client,
    model=model_id,
    instructions="You are a helpful assistant. Use the tools you have access to for providing relevant answers.",
    tools=[
        KnowledgeSearchTool(vector_store_id="1234"),
        KnowledgeSearchTool(vector_store_id="5678", name="paper_search", description="Search research papers"),
        KnowledgeSearchTool(vector_store_id="1357", name="wiki_search", description="Search wiki pages"),
        # no need to register toolgroup, just pass in the server uri
        # all available tools will be used
        MCPTool(server_uri="http://localhost:8000/sse"),
        # can specify a subset of available tools
        MCPTool(server_uri="http://localhost:8000/sse", tool_names=["list_directory"]),
        MCPTool(server_uri="http://localhost:8000/sse", tool_names=["list_directory"]),
        # custom tool
        my_custom_tool,
    ]
)
```

## Test Plan
# What does this PR do?


## Test Plan
# What does this PR do?


## Test Plan
This commit is contained in:
Eric Huang 2025-03-26 11:14:40 -07:00
parent 39e094736f
commit 7027b537e0
22 changed files with 951 additions and 525 deletions
Download patch file Download diff file Expand all files Collapse all files

10
llama_stack/providers/remote/inference/vllm/vllm.py
View file

@ -45,7 +45,7 @@ from llama_stack.apis.inference import (
ToolPromptFormat,
)
from llama_stack.apis.models import Model, ModelType
from llama_stack.models.llama.datatypes import BuiltinTool, StopReason, ToolCall
from llama_stack.models.llama.datatypes import StopReason, ToolCall
from llama_stack.models.llama.sku_list import all_registered_models
from llama_stack.providers.datatypes import ModelsProtocolPrivate
from llama_stack.providers.utils.inference.model_registry import (
@ -110,6 +110,8 @@ def _convert_to_vllm_tools_in_request(tools: List[ToolDefinition]) -> List[dict]
for tool in tools:
properties = {}
compat_required = []
tool_name = tool.name
if tool.parameters:
for tool_key, tool_param in tool.parameters.items():
properties[tool_key] = {"type": tool_param.param_type}
@ -120,12 +122,6 @@ def _convert_to_vllm_tools_in_request(tools: List[ToolDefinition]) -> List[dict]
if tool_param.required:
compat_required.append(tool_key)
# The tool.tool_name can be a str or a BuiltinTool enum. If
# it's the latter, convert to a string.
tool_name = tool.tool_name
if isinstance(tool_name, BuiltinTool):
tool_name = tool_name.value
compat_tool = {
"type": "function",
"function": {