staged-implementation: tighten planner and orchestrator review artifact handling

plugins/staged-implementation/skills/orchestrator/SKILL.md

@@ -49,6 +49,7 @@ For each ready chunk:
    - Read the plan, checklist, prompt map, recent review notes, and relevant git/worktree state.
    - Identify the next ready chunk from the checklist and prompt map.
    - Verify the chunk has not already landed.
+   - Run `git status --short` and relevant `git log --oneline` checks to verify the chosen chunk has not already landed and that the review target matches the current worktree.
 
 2. Write the implementer prompt.
    - Produce one surgical prompt for that chunk only.
@@ -66,6 +67,7 @@ For each ready chunk:
    - Compare against the frozen plan, checklist, prompt, and declared scope.
    - Verify the implementer's self-audit claims against the diff.
    - Lead review with findings ordered by severity.
+   - Save or update review outcomes beside the companion plan/checklist when the run is maintaining staged workflow artifacts.
 
 5. Handle review outcome.
    - If contract ambiguity exists, stop and identify the exact missing decision.
@@ -86,7 +88,7 @@ For each ready chunk:
    - Use the chunk's proposed commit message when acceptable; otherwise write a one-line commit message with the chunk ID prefix when one exists.
 
 8. Continue.
-   - Update or report checklist/prompt-map state as appropriate.
+   - Update or report checklist, prompt-map, and review-outcome state as appropriate.
    - Choose the next ready chunk.
    - Stop when all chunks are complete, blocked, or no ready chunk remains.
 

plugins/staged-implementation/skills/planner/SKILL.md

@@ -26,6 +26,7 @@ Treat questions, observations, and suggestions as analysis-only unless the user
 - Do not ask the implementer to guess.
 - Default to skepticism about unstated assumptions, explicit scope boundaries, and concrete filenames, routes, states, dates, fields, and error codes.
 - Keep `update_plan` synced for substantial work, with exactly one step in progress.
+- Never revert unrelated dirty changes; ignore them unless they directly block the requested planner/reviewer task.
 
 ## Required Inputs
 
@@ -140,6 +141,7 @@ Use the prompt map to preserve sequencing for the planner/reviewer. Do not inclu
 Before writing the prompt:
 
 - verify the correct next chunk from the checklist, prompt map, recent git history, and current worktree state when available
+- use `git status --short` and relevant `git log --oneline` checks before claiming a chunk is next, landed, or ready for review
 - state which chunk you chose and why
 - stop if an ambiguity blocks an implementer-quality prompt
 - include current-state context only to the extent needed for the implementer to execute the chunk without relying on prior chat memory
@@ -210,9 +212,22 @@ Before presenting or saving a prompt:
 - put `git log` options before any `--` pathspec separator
 - include `git diff --check` for prompts that may edit docs or code, unless a no-diff-check reason is explicit
 - when companion docs may be untracked, require `git status --short` plus readback or explicit untracked-file verification
+- if validation commands allow adjusted paths or equivalent local substitutions, require the implementer to report the exact adjustment in the final summary
 - keep scope, out-of-scope work, validation, test posture, output path, and save behavior aligned with the user request
 - do not add a visible prompt-lint report unless asked
 
+## Durable Review Artifacts
+
+When the staged workflow is maintaining on-disk artifacts, do not leave review outcomes only in chat. Save or update review notes or outcomes beside the companion plan/checklist unless the user explicitly asks for chat-only review.
+
+A review outcome should capture:
+
+- chunk ID/name
+- verdict: accepted, needs follow-up, blocked, or deferred
+- findings or explicit no-findings result
+- validation status and residual risk
+- next recommended action
+
 ## Review Rules
 
 When asked to review staged work:
@@ -249,6 +264,7 @@ Before ending a planner/reviewer turn, verify:
 - official prompts are in the companion plan/checklist directory unless another path was requested
 - filenames follow local feature naming conventions when such conventions are discoverable
 - saved official prompt contents were re-read
+- review outcomes were saved or updated when the workflow expects durable review artifacts
 - required workflow artifacts were not left only in chat
 - the next chunk can be described from docs and handoff artifacts without relying on prior session memory