Loom 🧵
loom is a high-performance, graph-based AI workflow engine for Go. It allows you to build complex agentic systems by defining multi-step workflows as directed graphs, seamlessly connecting them to any large language model, and persisting execution state for long-running or interrupted tasks.
Unlike linear chains, loom provides a robust, state-first architecture inspired by state machines, making it easy to build agents that can loop, branch, and resume exactly where they left off.
Key Features
- 🕸️ Graph-based Workflows: Define agents as directed graphs with nodes and conditional edges.
- 💾 State Persistence: Built-in checkpointing with support for PostgreSQL and SQLite.
- 📺 Real-time Streaming: First-class support for token-level streaming and event-based updates.
- 🧠 Advanced Memory Management: Intelligently condense long conversations and trim message history.
- 🔌 Provider Agnostic: One interface for OpenAI, Anthropic, Ollama, and Google Gemini.
- 🔌 MCP Support: Native integration with the Model Context Protocol for tools, resources, and prompts.
- 🛠️ Tool Integration: Native support for tool calling and tool-use loops.
- 🛡️ Type Safe: Built from the ground up with Go generics for maximum developer productivity.
Why Loom?
Loom is designed for developers who need more control than what simple chains or "black-box" agent frameworks offer. By treating your AI workflow as a state machine, Loom gives you:
- Explicit Control: You define exactly how the state transitions between nodes.
- Durability: Every step is checkpointed, so your agents can survive crashes or wait for human input.
- Observability: Built-in tools to visualize your graphs and trace every token and tool call.