README
¶
🤖 Docker cagent 🤖
Build, run, and share AI agents with a declarative YAML config, rich tool ecosystem, and multi-agent orchestration.
What is Docker cagent?
Docker cagent lets you create and run intelligent AI agents that collaborate to solve complex problems — no code required. Define agents in YAML, give them tools, and let them work.
agents:
root:
model: openai/gpt-5-mini
description: A helpful AI assistant
instruction: |
You are a knowledgeable assistant that helps users with various tasks.
Be helpful, accurate, and concise in your responses.
toolsets:
- type: mcp
ref: docker:duckduckgo
cagent run agent.yaml
Key Features
- Multi-agent architecture — Create teams of specialized agents that delegate tasks automatically
- Rich tool ecosystem — Built-in tools + any MCP server (local, remote, or Docker-based)
- AI provider agnostic — OpenAI, Anthropic, Gemini, AWS Bedrock, Mistral, xAI, Docker Model Runner, and more
- YAML configuration — Declarative, versionable, shareable
- Advanced reasoning — Built-in think, todo, and memory tools
- RAG — Pluggable retrieval with BM25, embeddings, hybrid search, and reranking
- Package & share — Push agents to any OCI registry, pull and run them anywhere
Install
Docker Desktop (4.49+) — cagent is pre-installed. Just run cagent.
Homebrew — brew install cagent
Binary releases — Download from GitHub Releases.
Set at least one API key (or use Docker Model Runner for local models):
export OPENAI_API_KEY=sk-... # or ANTHROPIC_API_KEY, GOOGLE_API_KEY, etc.
Quick Start
# Run the default agent
cagent run
# Run from the agent catalog
cagent run agentcatalog/pirate
# Generate a new agent interactively
cagent new
# Run your own config
cagent run agent.yaml
More examples in the examples/ directory.
Documentation
- Installation · Quick Start
- Agents · Models · Tools · Multi-Agent
- Configuration Reference
- TUI · CLI · MCP Mode · RAG
- Model Providers · Docker Model Runner
Contributing
Read the Contributing guide to get started. We use cagent to build cagent:
cagent run ./golang_developer.yaml
Telemetry
We collect anonymous usage data to improve the tool. See Telemetry.
Community
Documentation
¶
There is no documentation for this package.
Source Files
Directories
¶
| Path | Synopsis |
|---|---|
|
cmd
|
|
|
examples
|
|
|
golibrary/builtintool
command
|
|
|
golibrary/multi
command
|
|
|
golibrary/simple
command
|
|
|
golibrary/stream
command
|
|
|
golibrary/tool
command
|
|
|
gen
|
|
|
pkg
|
|
|
app/export
Package export provides HTML export functionality for cagent sessions.
|
Package export provides HTML export functionality for cagent sessions. |
|
audio/capture
Package capture provides audio capture functionality.
|
Package capture provides audio capture functionality. |
|
audio/transcribe
Package transcribe provides real-time audio transcription using OpenAI's Realtime API.
|
Package transcribe provides real-time audio transcription using OpenAI's Realtime API. |
|
connectrpc
Package connectrpc provides a Connect-RPC server implementation for the cagent API.
|
Package connectrpc provides a Connect-RPC server implementation for the cagent API. |
|
creator
Package creator provides functionality to create agent configurations interactively.
|
Package creator provides functionality to create agent configurations interactively. |
|
evaluation
Package evaluation provides an evaluation framework for testing agents.
|
Package evaluation provides an evaluation framework for testing agents. |
|
fake
Package fake provides a VCR-based proxy for replaying recorded AI API responses.
|
Package fake provides a VCR-based proxy for replaying recorded AI API responses. |
|
hooks
Package hooks provides lifecycle hooks for agent tool execution.
|
Package hooks provides lifecycle hooks for agent tool execution. |
|
model/provider/rulebased
Package rulebased provides a rule-based model router that selects the appropriate model based on NLP analysis of the input using Bleve.
|
Package rulebased provides a rule-based model router that selects the appropriate model based on NLP analysis of the input using Bleve. |
|
permissions
Package permissions provides tool permission checking based on configurable Allow/Ask/Deny patterns.
|
Package permissions provides tool permission checking based on configurable Allow/Ask/Deny patterns. |
|
sessiontitle
Package sessiontitle provides session title generation using a one-shot LLM call.
|
Package sessiontitle provides session title generation using a one-shot LLM call. |
|
tools/a2a
Package a2a provides a toolset implementation for connecting to remote A2A agents.
|
Package a2a provides a toolset implementation for connecting to remote A2A agents. |
|
tui/animation
Package animation provides centralized animation tick management for the TUI.
|
Package animation provides centralized animation tick management for the TUI. |
|
tui/cmdbatch
Package cmdbatch provides a fluent builder for batching tea.Cmd values.
|
Package cmdbatch provides a fluent builder for batching tea.Cmd values. |
|
tui/components/markdown
Package markdown provides a high-performance markdown renderer for terminal output.
|
Package markdown provides a high-performance markdown renderer for terminal output. |
|
tui/components/scrollview
Package scrollview provides a composable scrollable view that pairs content with a fixed-position scrollbar.
|
Package scrollview provides a composable scrollable view that pairs content with a fixed-position scrollbar. |
|
tui/messages
Package messages defines all TUI message types organized by domain.
|
Package messages defines all TUI message types organized by domain. |
|
tui/subscription
Package subscription provides patterns for external event sources in the TUI.
|
Package subscription provides patterns for external event sources in the TUI. |
|
userconfig
Package userconfig provides user-level configuration for cagent.
|
Package userconfig provides user-level configuration for cagent. |
Click to show internal directories.
Click to hide internal directories.