Snapshot and compare tool-call response contracts across configured LLM endpoints, with offline fixture replay and CI-readable drift reports.
-
Updated
Sep 8, 2026 - Python
Snapshot and compare tool-call response contracts across configured LLM endpoints, with offline fixture replay and CI-readable drift reports.
List recent model-repository updates, inspect local hardware and assemble static candidate configurations; live inference probing is not implemented.
Validate declarative tool-scope test batteries and classify recorded actions against allowlists; live model execution is not yet implemented.
Render Jinja chat templates and compare role markers and tool-call fields against repository-defined conformance specifications.
To associate your repository with the cn-models topic, visit your repo's landing page and select "manage topics."