open-multi-agent
TypeScript multi-agent framework that runs in your own environment: consequential actions wait for approval and every run leaves a verifiable record. Describe the goal, not the graph. 13 built-in providers (Claude, OpenAI, Gemini, DeepSeek and more) plus any OpenAI-compatible endpoint, local models included.
- stars
- 6,934
- 30-day movement
- starts with the next reading
- Related entries
- 61
- Connections
- 3
open-multi-agent is a TypeScript multi-agent framework that runs self-hosted in your own environment. It lets you describe goals rather than wiring agent graphs, gates consequential actions behind approval, records every run for verification, and supports 13 built-in providers plus any OpenAI-compatible endpoint, including local models.
You want multi-agent orchestration with approval controls and verifiable run records without sending work outside your own environment.
Use it to
- Describe goals and let agents coordinate without hand-building graphs
- Require human approval before consequential actions execute
- Verify what a run did via its recorded record
- Run agents against local models or OpenAI-compatible endpoints
For TypeScript developers building governed, self-hosted multi-agent systems
- Role
- agent-framework
- Language
- TypeScript
- Licence
- MIT
- Forks
- 2,432
- Open issues
- 4
- Last push
- 2026-09-14
- Latest release
- v1.0.0 · 2026-04-05