Review an implementation plan for completeness, feasibility, dependency coverage, and risk assessment.
Scanned 10/6/2026
npx -y skills add tomzx/agents --skill review-plan --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Review Plan?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/tomzx-review-plan)More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.
---
name: review-plan
description: Review an implementation plan for completeness, feasibility, dependency coverage, and risk assessment.
---
# Review Plan
Reviews an implementation plan and reports findings across six categories: completeness, feasibility, dependencies, risk coverage, timeline realism, and reversibility.
## Prerequisites
- Apply the shared SDLC conventions in `skills/sdlc/references/shared.md`.
- If no argument is provided, locate the feature directory under `.sdlc/features/` whose frontmatter `issue` field references `$ISSUE_NUMBER`.
- `.sdlc/features/N-<slug>/plan.md` (unified) **or** `.sdlc/features/N-<slug>/plan/index.md` plus its sibling `plan/<concern>.md` files (split), or an implementation plan provided in context or as a file path
- `.sdlc/features/N-<slug>/specification.md` (optional, improves coverage analysis)
## Steps
1. **Resolve the plan.** Look for `.sdlc/features/N-<slug>/plan.md` first; if absent, look for `.sdlc/features/N-<slug>/plan/index.md` and read it together with every `plan/<concern>.md` it lists. Otherwise read from context or as a file path. Treat the whole plan set (index + concern files) as the unit under review.
2. Cross-reference against the specification or requirements if available.
3. Run the deterministic checker when possible: render each `mermaid` block with `mmdc` (or `npx -y @mermaid-js/mermaid-cli`) when available. A tool that is not installed is skipped (never blocks); a render failure is a blocking finding under Dependencies or Timeline Realism.
4. Identify issues in each category below. For a split plan, also check that `plan/index.md` aggregates milestones, cross-concern dependencies, risks, and timeline consistently with the concern files.
5. Report findings. Omit any category that has no findings.
6. Write the findings to `.sdlc/features/N-<slug>/review-plan.md` with frontmatter `artifact: plan`, `verdict` (`approved` if there are no blocking findings, `changes-requested` if the author must address findings, `rejected` for a fundamental flaw), and `reviewed_at: <ISO date>`, and the findings as the body, per `skills/sdlc/references/shared.md`. Record any unresolved open questions in the findings body. For any question that carries meaningful risk to the implementation, also invoke `/create-assumption` to record it formally.
## Review Checklist
### Completeness
- Does the plan cover all requirements and spec deliverables?
- Are all phases clearly defined with success criteria?
- Are setup, deployment, and rollout steps included?
### Feasibility
- Are effort estimates realistic for the described work?
- Does the plan account for ramp-up, reviews, and integration work?
- Are milestones achievable within the stated constraints?
### Dependencies
- Are all internal and external dependencies identified?
- Does the phase-dependency flowchart match the per-phase `Depends on:` fields (no edge missing, no edge invented)?
- Are critical-path dependencies clearly marked?
- Is there a contingency for delayed or unavailable dependencies?
### Risk Coverage
- Are the most significant risks captured in the risk register?
- Does each risk have a concrete mitigation strategy?
- Are there single points of failure not mentioned as risks?
### Timeline Realism
- Is the timeline consistent with the effort estimates (gantt durations vs. phase effort)?
- When a gantt is present, do its task dependencies match the phase-dependency flowchart?
- Are there parallel tracks that could shorten total duration?
- Are buffer periods included for testing and review?
### Reversibility
- Can we undo this cleanly once implemented, or does the plan create one-way-door commitments?
- Does the plan include a rollback path for each phase (migrations, deployments, config)?
- Are irreversible steps (destructive migrations, deletions, public API removals) flagged and ordered safely?
## Output Format
```markdown
## Completeness
<Findings or "No issues found.">
## Feasibility
<Findings or "No issues found.">
## Dependencies
<Findings or "No issues found.">
## Risk Coverage
<Findings or "No issues found.">
## Timeline Realism
<Findings or "No issues found.">
## Reversibility
<Findings or "No issues found.">
```
## Outcome
If `$OUTCOME_YAML` is set, emit your verdict there per `skills/sdlc/references/shared.md`:
| Verdict | When |
|---|---|
| `approved` | No blocking findings; the subject passes review |
| `changes-requested` | Findings the author must address before it passes |
| `rejected` | Fundamental flaw requiring rework or stopping |
In the same emission, list the findings file under `artifacts:` (`.sdlc/features/N-<slug>/review-plan.md`).
## Example Usage
**Scenario 1: Missing rollout step**
Plan ends at "integration testing complete" with no deployment or rollout phase.
Report under Completeness.
**Scenario 2: Underestimated effort**
Phase 2 (API + auth) is estimated at 1 day for a spec that describes 8 endpoints with complex permission logic.
Report under Feasibility.
**Scenario 3: Unmitigated critical dependency**
Plan depends on a third-party API but lists no spike or contingency if that API is unavailable.
Report under Risk Coverage.
## Next Step
Once the findings verdict is `approved`, run `/publish-plan` to commit the plan and open a draft PR for author sign-off, then continue with `/create-tasks-decomposition`.
## Useful Commands Reference
| Command | Description |
|---|---|
| `mmdc -i <diagram.mmd>` or `npx -y @mermaid-js/mermaid-cli` | Best-effort Mermaid render check; failure is a blocking finding |
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!