sd plan review <pl-id> --by <name> — record a reviewer’s name against a plan. “Informational; not a state transition.”
It is listed here for completeness and because of the gap it exposes. The framework’s thesis is that plans should be the artifact reviewed — attention spent before implementation rather than after — and this is the entire mechanism supporting it: a free-text name field, plus a “review suggested” hint that seeds-plan-show prints when a plan is approved or active with no reviewer recorded. Nothing gates on it, nothing checks that a review happened, and the reviewer records no findings.
So it maps to no canonical SDLC stage: storing who is said to have looked at a plan does not perform stage-review in the sense the other frameworks’ review capabilities do — gstack-plan-eng-review and ce-doc-review actually read the plan and return findings. The honest reading is that seeds asserts the review thesis and delegates the execution of it to whatever agent or human the team points at the plan; what seeds contributes is the validation gate (seeds-plan-submit), which is a different control entirely.
See Also
- seeds-plan-show — prints the review-suggested hint.
- seeds-plan-submit — the framework’s real, deterministic plan gate.
- gstack-plan-eng-review · ce-doc-review · gsd-plan-checker — plan reviews that produce findings.
- stage-review — the stage this deliberately does not implement.