Run these from the root of your Python agent repo.
Paste this into your coding agent
Or install yourself, then ask your agent
If you want to control the install step but still let your coding agent drive the eval workflow. For uv projects:requirements.txt, add a line:
kensa-evals skill. With a trace source, kensa init stores it and the redaction model under [tool.kensa] and prepares redaction. Use --redaction-model large for higher recall.
kensa init scaffolds tests/evals/conftest.py (the kensa_run harness), a tests/evals/test_kensa_smoke.py smoke eval, a GitHub Actions workflow, and the coding-agent instructions.
Or CLI-only
If you want to drive Kensa directly:kensa import minimizes provider payloads to Kensa’s allowlisted TraceView, redacts retained values, and writes the evidence to .kensa/traces/imports/. Read it with kensa traces list/sample/get, then turn it into evals: the kensa-inspect skill proposes a YAML review queue you validate with kensa inspect lint, you approve ideas, and the kensa-generate skill materializes tests/evals/test_<id>.py. If you do not have traces yet, capture a local run with kensa.instrument() and import the JSONL, then run kensa eval.
What kensa init sets up
kensa init scaffolds the harness and installs kensa-evals. Setup follows repository source to
the production function or class that starts one conversation and waits for approval before writing
a kensa_run adapter. If it cannot call production code without recreating agent behavior, it
reports cannot wire.
See The harness fixture for the adapter example and
Setup safety for ownership and safety rules. kensa doctor verifies that the
adapter calls the approved production code before evidence collection begins.
Next
- Concepts - cases, traces, assertions, judges, trials
- Cases - define inputs and trials with
kensa_case - Tracing & imports -
kensa connect,kensa import, local capture - CLI reference - the full command surface