| name | start-review |
| description | Start a review session from an existing plan and implementation. Discovers available plans, validates implementation exists, and invokes the technical-review skill. |
| disable-model-invocation | true |
| allowed-tools | Bash(.claude/skills/start-review/scripts/discovery.sh), Bash(.claude/hooks/workflows/write-session-state.sh) |
| hooks | {"PreToolUse":[{"hooks":[{"type":"command","command":"$CLAUDE_PROJECT_DIR/.claude/hooks/workflows/system-check.sh","once":true}]}]} |
Invoke the technical-review skill for this conversation.
⚠️ ZERO OUTPUT RULE: Do not narrate your processing. Produce no output until a step or reference file explicitly specifies display content. No "proceeding with...", no discovery summaries, no routing decisions, no transition text. Your first output must be content explicitly called for by the instructions.
Workflow Context
This is Phase 6 of the six-phase workflow:
| Phase | Focus | You |
|---|
| 1. Research | EXPLORE - ideas, feasibility, market, business | |
| 2. Discussion | WHAT and WHY - decisions, architecture, edge cases | |
| 3. Specification | REFINE - validate into standalone spec | |
| 4. Planning | HOW - phases, tasks, acceptance criteria | |
| 5. Implementation | DOING - tests first, then code | |
| 6. Review | VALIDATING - check work against artifacts | ◀ HERE |
Stay in your lane: Verify that every plan task was implemented, tested adequately, and meets quality standards. Don't fix code - identify problems. You're reviewing, not building.
Instructions
Follow these steps EXACTLY as written. Do not skip steps or combine them. Present output using the EXACT format shown in examples - do not simplify or alter the formatting.
CRITICAL: This guidance is mandatory.
- After each user interaction, STOP and wait for their response before proceeding
- Never assume or anticipate user choices
- Even if the user's initial prompt seems to answer a question, still confirm with them at the appropriate step
- Complete each step fully before moving to the next
- Do not act on gathered information until the skill is loaded - it contains the instructions for how to proceed
Step 0: Run Migrations
This step is mandatory. You must complete it before proceeding.
Invoke the /migrate skill and assess its output.
If files were updated: STOP and wait for the user to review the changes (e.g., via git diff) and confirm before proceeding to Step 1. Do not continue automatically.
If no updates needed: Proceed to Step 1.
Step 1: Discovery State
!.claude/skills/start-review/scripts/discovery.sh
If the above shows a script invocation rather than YAML output, the dynamic content preprocessor did not run. Execute the script before continuing:
.claude/skills/start-review/scripts/discovery.sh
If YAML content is already displayed, it has been run on your behalf.
Parse the discovery output to understand:
From plans section:
exists - whether any plans existfiles - list of plans with: name, topic, status, date, format, specification, specification_exists, plan_id (if present)count - total number of plans
From reviews section:
exists - whether any reviews existentries - list of reviews with: scope, type, plans, versions, latest_version, latest_verdict, latest_path, has_synthesis
From state section:
scenario - one of: "no_plans", "single_plan", "multiple_plans"implemented_count - plans with implementation_status != "none"completed_count - plans with implementation_status == "completed"reviewed_plan_count - plans that have been reviewedall_reviewed - whether all implemented plans have reviews
IMPORTANT: Use ONLY this script for discovery. Do NOT run additional bash commands (ls, head, cat, etc.) to gather state - the script provides everything needed.
→ Proceed to Step 2.
Step 2: Route Based on Scenario
Use state.scenario from the discovery output to determine the path:
If scenario is "no_plans"
No plans exist yet.
Output the next fenced block as a code block:
Review Overview
No plans found in .workflows/planning/
The review phase requires a completed implementation based on a plan.
Run /start-planning first to create a plan, then /start-implementation
to build it.
STOP. Do not proceed — terminal condition.
If all_reviewed is true
All implemented plans have been reviewed.
Output the next fenced block as a code block:
Review Overview
All {N} implemented plans have been reviewed.
1. {topic:(titlecase)}
└─ Review: x{review_count} — r{latest_review_version} ({latest_review_verdict})
└─ Synthesis: @if(has_synthesis) completed @else pending @endif
2. ...
Output the next fenced block as markdown (not a code block):
· · · · · · · · · · · ·
All plans have been reviewed.
- **`a`/`analysis`** — Synthesize findings from existing reviews into tasks
- **`r`/`re-review`** — Re-review a plan (creates new review version)
Select an option:
· · · · · · · · · · · ·
STOP. Wait for user response.
If analysis
→ Proceed to Step 4 with scope set to "analysis".
If re-review
→ Proceed to Step 3, incrementing the review version for the selected plan.
If scenario is "single_plan" or "multiple_plans"
Plans exist (some may have reviews, some may not).
→ Proceed to Step 3 to present options.
Step 3: Display Plans
Load display-plans.md and follow its instructions as written.
→ Proceed to Step 4.
Step 4: Select Plans
Load select-plans.md and follow its instructions as written.
→ Proceed to Step 5.
Step 5: Invoke the Skill
Load invoke-skill.md and follow its instructions as written.
