OpenWeave

OpenWeave is execution governance for AI agents and autonomous systems — a server-enforced state machine that controls what AI agents can do and when. You define workflows as explicit states, the transitions allowed between them, and exactly who may trigger each one (humans, specific bots, or both). The backend enforces every transition with a hard error: if an agent attempts a move that isn't permitted from the current state, the API returns a 403 and the transition does not happen.

Critical states can sit behind human approval gates, so a bot is blocked until a human signs off — deterministic human-in-the-loop control where it matters, and full autonomy everywhere else. Monitoring tools tell you what an agent did, after the fact; OpenWeave prevents what agents shouldn't do, before it happens.

Every agent has a verifiable identity, agents discover allowed transitions from the API instead of hardcoding them, and every change is written to an immutable audit trail. It integrates over a REST API and a remote MCP server, dropping into existing agent frameworks and LLM tooling. Built for AI-agent and autonomous-systems developers, AgentOps/MLOps and platform teams.

Freemium: a free tier plus a $29/mo Pro plan.