Playbook
Review a design
Paste the design. Score it against named failures and the five readiness questions.
When: A design doc, a PR that adds an agent, or a framework pitch.
- 01
Run the reviewer
Paste into /review or POST /api/canon/review. Do not review by vibe.
The Named Failures - 02
The five questions
Termination, sandbox, context builder, eval set, pause API. Missing is a block, not a nit.
Anti-Patterns - 03
Done when
- Every hit has a named failure and a next article. Ready-to-build is either green or a ticket.
Refuse
- Eval by demo. 20–50 real tasks before the framework. promptfoo, Inspect, DSPy. CI on prompt change.
- Unbounded loop. Force a done schema, a step cap, and a money cap. 'Done' is a tool or a structured intent.
- Unsandboxed exec. E2B, a container, OpenHands isolation, or a policy sandbox. Secrets stay in the executor.
- Star-driven selection. Pick for the constraint: checkpoints, types, harness, documents. Read Canon's atlas.