docket-runtime is an open-source CLI tool designed for triaging and observing LLM agent traces in a platform-agnostic way. It supports integration with multiple observability platforms, enables failure mode analysis, and implements the MCP protocol for interoperability. Ideal for AI engineers and developers working on agent observability.
docket-runtime sits in PulseGate's LLM eval & observability category. It focuses on diagnosing and triaging LLM agent failures across different observability platforms. docket-runtime is an open-source project aimed at AI engineers and observability platform developers. The project is open source (Apache-2.0). docket-runtime is available on the command line.
It is developed by wczaja, and the product first shipped in 2026. The project is developed in the open on GitHub with 27 commits in the last 90 days. Across PulseGate's embedding index, docket-runtime has few near neighbours, marking it as relatively distinct. Among its 5 catalogued features are agent trace triage, observability integration, and failure mode analysis. It exposes integrations via an MCP server.
Latest indexed changes and source events
docket-runtime discovered by the PulseGate indexer
Other apps tracked under the same category.
.png){kind=small}
