Back to directory

LightAgent

An ultra-lightweight, open-source Python framework for building OpenAI-compatible agents, with tools, persistent memory, guardrails, tracing, multi-agent collaboration and workflows.

LightAgent

LightAgent is an open-source Python framework for building agents that work with any modern LLM, including OpenAI, DeepSeek and Qwen models, and expose an OpenAI-compatible streaming API so they can be dropped into an existing chat interface. It natively supports Skills, letting reusable capabilities be composed with persistent memory, tool use, and tree-of-thought reasoning, and connects to MCP servers over stdio and SSE.

It streamlines multi-agent collaboration, and its runtime adds an event-sourced Agent Runtime with durable sessions, asynchronous execution, a capability registry and policy layer, job and subagent handling, and SQLite full-text-search-based retrieval. Recent releases have added a dependency-free connector contract, Python-executor security checks, a security matrix for shared graph memory, checkpointed workflow orchestration with resume and rerun support, and approval nodes for human oversight.

It is released under the Apache-2.0 license.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.