Atl
Execute supported agent tasks through verified providers with fallback and durable outcomes.
- Transport
- Not stated
- Package
- —
- Registry id
- com.agenttrafficlab/atl
No install snippet on purpose. A working MCP config is a command, its arguments and an environment block — the last two are where API keys live, so this catalogue never stores them and cannot publish them. Follow the link above for the authors' own instructions.
Agent Traffic Lab routes supported AI tasks to eligible providers, executes them with bounded fallback, and records durable outcomes. The internal Decision latency is approximately 1 millisecond in production. Serialized Decision response bodies are 91.3% smaller using compact mode.
Users should use atl_complete_task for end-to-end execution so that ATL handles routing, fallback, and outcome recording. The product integrates via MCP at the endpoint mcp.agenttrafficlab.com. The Python routing client has no runtime dependencies.
The JavaScript routing client requires Node.js version 18 or higher.
Summary drafted from the project's own website. Every sentence is backed by text on that page and was reviewed before publishing.