[IMPORTANT] Use TaskCreate to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI may ask user whether to skip.
Prerequisites: MUST READ .claude/skills/shared/understand-code-first-protocol.md before executing.
Critical Purpose: Ensure quality — no flaws, no bugs, no missing updates, no stale content. Verify both code AND documentation.
Quick Summary
Goal: Auto-review implementation plans for validity, correctness, and best practices before proceeding (AI self-review, not user interview).
Workflow:
- Resolve Plan — Use $ARGUMENTS path or active plan from
## Plan Context
- Read Files — plan.md + all phase-*.md files, extract requirements/steps/files/risks
- Evaluate Checklist — Validity (summary, requirements, steps, files), Correctness (specific, paths, no conflicts), Best Practices (YAGNI/KISS/DRY, architecture), Completeness (risks, testing, success, security)
- Score & Classify — PASS (all Required + ≥50% Recommended), WARN (all Required + <50% Recommended), FAIL (any Required fails)
- Output Result — Status, checks passed, issues, recommendations, verdict
Key Rules:
- PASS: Proceed to implementation
- WARN: Proceed with caution, note gaps
- FAIL: STOP - must fix before proceeding, list specific issues
- Constructive: Focus on implementation-blocking issues, not pedantic details
Your mission
Perform automatic self-review of an implementation plan to ensure it's valid, correct, follows best practices, and identify anything needing fixes before proceeding.
Key distinction: This is AI self-review (automatic), NOT user interview like /plan-validate.
Plan Resolution
- If
$ARGUMENTS provided -> Use that path
- Else check
## Plan Context section -> Use active plan path
- If no plan found -> Error: "No plan to review. Run /plan first."
Workflow
Step 1: Read Plan Files
Read the plan directory:
plan.md - Overview, phases list, frontmatter
phase-*.md - All phase files
- Extract: requirements, implementation steps, file listings, risks
Step 2: Evaluate Against Checklist
Validity (Required - all must pass)
Correctness (Required - all must pass)
Best Practices (Required - all must pass)
Completeness (Recommended - ≥50% should pass)
Step 3: Score and Classify
| Status |
Criteria |
Action |
| PASS |
All Required pass, ≥50% Recommended |
Proceed to implementation |
| WARN |
All Required pass, <50% Recommended |
Proceed with caution, note gaps |
| FAIL |
Any Required check fails |
STOP - must fix before proceeding |
Step 4: Output Result
## Plan Review Result
**Status:** PASS | WARN | FAIL
**Reviewed:** {plan-path}
**Date:** {current-date}
### Summary
{1-2 sentence summary of plan quality}
### Checks Passed ({X}/{Y})
#### Required ({X}/{Y})
- ✅ Check 1
- ✅ Check 2
- ❌ Check 3 (if failed)
#### Recommended ({X}/{Y})
- ✅ Check 1
- ⚠️ Check 2 (missing)
### Issues Found
- ❌ FAIL: {critical issue requiring fix}
- ⚠️ WARN: {minor issue, can proceed}
### Recommendations
1. {specific fix 1}
2. {specific fix 2}
### Verdict
{PROCEED | REVISE_FIRST | BLOCKED}
Next Steps
- If PASS: Announce "Plan review complete. Proceeding with next workflow step."
- If WARN: Announce "Plan review complete with warnings. Proceeding - consider addressing gaps."
- If FAIL: List specific issues. Do NOT proceed. Ask user to fix or regenerate plan.
IMPORTANT Task Planning Notes (MUST FOLLOW)
- Always plan and break work into many small todo tasks using
TaskCreate
- Always add a final review todo task to verify work quality and identify fixes/enhancements
- MANDATORY FINAL TASKS: After creating all planning todo tasks, ALWAYS add these two final tasks:
- Task: "Run /plan-validate" — Trigger
/plan-validate skill to interview the user with critical questions and validate plan assumptions
- Task: "Run /plan-review" — Trigger
/plan-review skill to auto-review plan for validity, correctness, and best practices
Important Notes
- Be constructive, not pedantic - focus on issues that would cause implementation problems
- WARN is acceptable for missing optional sections
- FAIL only for genuinely missing required content
- If plan is simple and valid, quick review is fine
1---2name: plan-review-73description: [Planning] Auto-review plan for validity, correctness, and best practices before implementation4---5
6> **[IMPORTANT]** Use `TaskCreate` to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI may ask user whether to skip.
7
8**Prerequisites:** **MUST READ** `.claude/skills/shared/understand-code-first-protocol.md` before executing.
9
10> **Critical Purpose:** Ensure quality — no flaws, no bugs, no missing updates, no stale content. Verify both code AND documentation.
11
12## Quick Summary
13
14**Goal:** Auto-review implementation plans for validity, correctness, and best practices before proceeding (AI self-review, not user interview).
15
16**Workflow:**
17
181. **Resolve Plan** — Use $ARGUMENTS path or active plan from `## Plan Context`
192. **Read Files** — plan.md + all phase-\*.md files, extract requirements/steps/files/risks
203. **Evaluate Checklist** — Validity (summary, requirements, steps, files), Correctness (specific, paths, no conflicts), Best Practices (YAGNI/KISS/DRY, architecture), Completeness (risks, testing, success, security)
214. **Score & Classify** — PASS (all Required + ≥50% Recommended), WARN (all Required + <50% Recommended), FAIL (any Required fails)
225. **Output Result** — Status, checks passed, issues, recommendations, verdict
23
24**Key Rules:**
25
26- **PASS**: Proceed to implementation
27- **WARN**: Proceed with caution, note gaps
28- **FAIL**: STOP - must fix before proceeding, list specific issues
29- **Constructive**: Focus on implementation-blocking issues, not pedantic details
30
31## Your mission
32
33Perform automatic self-review of an implementation plan to ensure it's valid, correct, follows best practices, and identify anything needing fixes before proceeding.
34
35**Key distinction**: This is AI self-review (automatic), NOT user interview like `/plan-validate`.
36
37## Plan Resolution
38
391. If `$ARGUMENTS` provided -> Use that path
402. Else check `## Plan Context` section -> Use active plan path
413. If no plan found -> Error: "No plan to review. Run /plan first."
42
43## Workflow
44
45### Step 1: Read Plan Files
46
47Read the plan directory:
48
49- `plan.md` - Overview, phases list, frontmatter
50- `phase-*.md` - All phase files
51- Extract: requirements, implementation steps, file listings, risks
52
53### Step 2: Evaluate Against Checklist
54
55#### Validity (Required - all must pass)
56
57- [ ] Has executive summary (clear 1-2 sentence description)
58- [ ] Has defined requirements section
59- [ ] Has implementation steps (actionable tasks)
60- [ ] Has files to create/modify listing
61
62#### Correctness (Required - all must pass)
63
64- [ ] Steps are specific and actionable (not vague)
65- [ ] File paths follow project patterns
66- [ ] No conflicting or duplicate steps
67- [ ] Dependencies between steps are clear
68
69#### Best Practices (Required - all must pass)
70
71- [ ] YAGNI: No unnecessary features or over-engineering
72- [ ] KISS: Simplest viable solution chosen
73- [ ] DRY: No planned duplication of logic
74- [ ] Architecture: Follows project patterns from `docs/claude/`
75
76#### Completeness (Recommended - ≥50% should pass)
77
78- [ ] Risk assessment present with mitigations
79- [ ] Testing strategy defined
80- [ ] Success criteria per phase
81- [ ] Security considerations addressed
82
83### Step 3: Score and Classify
84
85| Status | Criteria | Action |
86| -------- | ----------------------------------- | --------------------------------- |
87| **PASS** | All Required pass, ≥50% Recommended | Proceed to implementation |
88| **WARN** | All Required pass, <50% Recommended | Proceed with caution, note gaps |
89| **FAIL** | Any Required check fails | STOP - must fix before proceeding |
90
91### Step 4: Output Result
92
93```markdown
94## Plan Review Result
95
96**Status:** PASS | WARN | FAIL
97**Reviewed:** {plan-path}
98**Date:** {current-date}
99
100### Summary
101
102{1-2 sentence summary of plan quality}
103
104### Checks Passed ({X}/{Y})
105
106#### Required ({X}/{Y})
107
108- ✅ Check 1
109- ✅ Check 2
110- ❌ Check 3 (if failed)
111
112#### Recommended ({X}/{Y})
113
114- ✅ Check 1
115- ⚠️ Check 2 (missing)
116
117### Issues Found
118
119- ❌ FAIL: {critical issue requiring fix}
120- ⚠️ WARN: {minor issue, can proceed}
121
122### Recommendations
123
1241. {specific fix 1}
1252. {specific fix 2}
126
127### Verdict
128
129{PROCEED | REVISE_FIRST | BLOCKED}
130```
131
132## Next Steps
133
134- **If PASS**: Announce "Plan review complete. Proceeding with next workflow step."
135- **If WARN**: Announce "Plan review complete with warnings. Proceeding - consider addressing gaps."
136- **If FAIL**: List specific issues. Do NOT proceed. Ask user to fix or regenerate plan.
137
138## **IMPORTANT Task Planning Notes (MUST FOLLOW)**
139
140- Always plan and break work into many small todo tasks using `TaskCreate`
141- Always add a final review todo task to verify work quality and identify fixes/enhancements
142- **MANDATORY FINAL TASKS:** After creating all planning todo tasks, ALWAYS add these two final tasks:
143 1. **Task: "Run /plan-validate"** — Trigger `/plan-validate` skill to interview the user with critical questions and validate plan assumptions
144 2. **Task: "Run /plan-review"** — Trigger `/plan-review` skill to auto-review plan for validity, correctness, and best practices
145
146## Important Notes
147
148- Be constructive, not pedantic - focus on issues that would cause implementation problems
149- WARN is acceptable for missing optional sections
150- FAIL only for genuinely missing required content
151- If plan is simple and valid, quick review is fine