Verify the project's riskiest invariant with TLA+, Lean, or Kani — chosen by property shape — and leave a project-local skill the next agent reruns; assumptions are part of the deliverable. For properties tests cannot reach.
Adversarial pass over a diff or design: three reviewers on different models each try to break it — correctness, security, edge cases — findings merged by severity into a blocking verdict. For contested or high-risk changes.
Spec-Driven Development. Interview → spec with acceptance criteria → plan → test-first execution. `/sdd full` runs the 6-phase pipeline instead.
Self-referential loop until task completion with configurable verification reviewer
Clean AI-generated code slop with a regression-safe, deletion-first workflow and optional reviewer-only mode
Settle types and module shape before cross-boundary code — competing designs explored in parallel, compared on API surface and invariants, decision recorded with rejected alternatives and an accepted-deviations register.
Run N parallel candidates at one brief and synthesize: rubric-first framing, isolated fan-out, a cross-judge on a different model, criterion-by-criterion pick, hand grafting. When one attempt would lock in the wrong shape.
Drive an open PR to merge-ready on the watch-pr verdict policy — watch checks and review state, fix CI, address comments; never merges, rebases, or force-pushes. For 'get this green' or 'babysit PR #N'.