Skip to content

Installation

Basic Installation

Simple installation

uv tool install llmling-agent[default]

Multiple extras

uv tool install llmling-agent[default, coding]

Available Extras

Name Dependencies Description
a2a fasta2a, starlette A2A Server
ag-ui ag-ui-protocol>=0.1.10 Ag-UI Agents
braintrust braintrust, autoevals Braintrust Prompt Hub
clipboard clipman Clipboard functionality
coding ast-grep-py>=0.40.0 Packages to allow coding functionality
composio composio Composio toolsets
default pydantic-ai-slim[openai,google,anthropic,mistral,retries]>=1.0.0 Set of commonly used AI providers
events evented[all] Event triggers for agents
git fsspec[git] Git support
langfuse langfuse Langfuse Prompt Hub
markitdown markitdown; python_version < '3.14' MarkItDown Media Converter
mcp_run mcpx-py>=0.7.0 MCP.run Toolset
memory typeagent>=0.3.3 TypeAgent semantic knowledge DB
notifications apprise>=1.9.5 Notification Toolset
oauth keyring>=25.6.0 OAuth functionality (will be fixed dep for FastMCP >2.13)
promptlayer promptlayer PromptLayer Prompt Hub
server logfire[fastapi], fastapi, uvicorn[standard] OpenAI Server functionality
textual textual>=1.0.0 Textual CLI app to create configs
tiktoken tiktoken Exact token counting
tree-sitter rustworkx>=0.17.1, grep-ast, tree-sitter>=0.25.2, tree-sitter-python>=0.25.0, tree-sitter-c>=0.24.1, tree-sitter-javascript>=0.25.0, tree-sitter>=0.25.2, tree-sitter-typescript>=0.23.0, tree-sitter-cpp>=0.23.0, tree-sitter-rust>=0.23.0, tree-sitter-go>=0.23.0, tree-sitter-json>=0.24.0, tree-sitter-yaml>=0.6.0 Tree-sitter based filesystem
tts openai>=1.0.0, sounddevice>=0.5.0, numpy>=1.24.0 Text-to-Speech with OpenAI
tts-edge edge-tts>=7.0.0, miniaudio>=1.60, sounddevice>=0.5.0 Text-to-Speech with Edge TTS (free, no API key)

One-Line ACP Setup

No installation needed - run directly with uvx:

uvx --python 3.13 llmling-agent[default]@latest serve-acp 

# or

uvx --python 3.13 llmling-agent[default]@latest serve-acp path/to/agents.yml