Back to directory

Gocrewwai

A strictly-typed, Go-native alternative to CrewAI for orchestrating parallel AI agents, with LangGraph-style durable flow persistence, agent-to-agent and MCP protocol support, human-in-the-loop approvals, and security-hardened tool execution.

Gocrewwai

Gocrewwai is a high-performance agentic orchestration framework for Go, inspired by CrewAI, LangChain and LangGraph, aimed at production use: goroutine-based concurrency gives true parallel agent execution, every LLM response is unmarshaled into strictly typed Go structs rather than handled as loosely typed data, and the whole orchestrator compiles into a single zero-dependency binary.

Durable flows checkpoint state to SQLite or Redis after every node so long-running workflows can pause, resume and time-travel; human-in-the-loop interrupts and approvals work from the CLI or a real-time dashboard; agents can reflect on and correct their own work through internal loops or peer-review crews; native OpenTelemetry tracing gives vendor-neutral observability; and it supports the Model Context Protocol with local and remote servers, and agent-to-agent protocols for decentralized agent swarms with peer health checks.

Code execution runs in hardened Docker sandboxes with no network, dropped capabilities, a read-only filesystem and a process limit; URL-fetching tools validate schemes and block private IP ranges against SSRF; shell commands are matched against an explicit whitelist that denies everything by default; dangerous tools such as file writes, HTTP requests, database operations and code execution require explicit human approval; and API authentication uses constant-time comparison with a request body size limit. Built with Go; MIT licensed; alpha and beta status heading toward a 1.0 release.

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.