Docker Agent
ActiveDescription
An AI Agent builder and runtime by Docker Engineering, bringing container-native isolation, portability, and standardization to AI agent lifecycle management from development through production deployment.
Key Features
- Multi-agent architecture - Create specialized agent teams with automatic task delegation and coordination
- Rich tool ecosystem - Built-in tools plus any MCP server (local, remote, or Docker-based)
- Multi-provider support - Compatible with OpenAI, Anthropic, Gemini, AWS Bedrock, Mistral, xAI, and Docker Model Runner for local models
- YAML declarative configuration - Versionable, shareable agent definitions
- Built-in advanced reasoning - Integrated think, todo, and memory tools
- RAG support - Pluggable retrieval with BM25, embeddings, hybrid search, and reranking
Use Cases
💡 Quickly build and run AI agents using declarative YAML config with no code required
💡 Package and distribute agents via OCI registries for standardized team sharing
💡 Handle complex tasks through multi-agent collaboration with automatic subtask decomposition
💡 Safely run agents in containerized environments leveraging Docker's isolation and portability
Categories
Quick Start
Install Docker Desktop 4.63+ (docker-agent plugin pre-installed) or use brew install docker-agent. Set an API key: export OPENAI_API_KEY=sk-..., then run docker agent run to start the default agent, docker agent run agentcatalog/pirate for a catalog agent, or docker agent new to create a new agent interactively.