Skip to content
cccskills
FreePublished on GitHub

review-plan

XP-style plan review that identifies gaps, hidden assumptions, and critical misses. Triggers on 'review the plan', 'check the plan', 'audit the plan'. Reviews plans for YAGNI violations, missing feedback loops, and over-engineering.

The English summary is being prepared. Showing the original description.

See how to install

Files included (2)

  • SKILL.md3.0 KB
  • workflow.js8.4 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

XP Plan Review

STARTER_CHARACTER = 👁️

Activation Triggers

  • "review the plan"
  • "/review-plan"
  • "check the plan"
  • "audit the plan"

Process

1. Find the Plan

Search for plan files:

  • Recent plan in context window user is referring to
  • .claude/plans/
  • plans/
  • docs/plans/
  • File mentioned by user

If multiple found, ask which one.

2. Run the Review Workflow

The multi-agent review — one independent reviewer per lens (Gaps, Hidden Assumptions, Critical Misses, plus the five XP Violations: YAGNI, Simplicity, Step Size, Feedback Loops, Tech Debt), dedup, and an adversarial-verify pass where skeptics vote to kill unconvincing findings — is encoded as a deterministic Workflow so the fan-out, dedup, and verification run reliably instead of relying on prose. The reviewers are Plan agents, exactly as before.

Invoke the Workflow tool with the bundled script, passing whichever plan reference you resolved in step 1 (path for a file, or plan for raw text):

Workflow({
  scriptPath: "${CLAUDE_PLUGIN_ROOT}/skills/review-plan/workflow.js",
  args: { path: "[PATH]" }   // or { plan: "[plan text]" }
})

The workflow runs two phases (both autonomous — no human gate):

  • Review (8 parallel Plan reviewers, independent): one per lens — Gaps in Knowledge · Hidden Assumptions · Critical Misses · YAGNI · Simplicity · Step Size · Feedback Loops · Tech Debt. Each returns findings with issue, impact, recommendation, and priority (critical > important > nice-to-have). Empty / issue-less findings are dropped; near-duplicate findings across lenses are merged.
  • Verify (adversarial skeptics, Haiku, one per unique finding): each skeptic tries to KILL the finding — assumes false positive until the plan's content proves otherwise. Findings the skeptic votes down or scores below 50 confidence are dropped.

It returns { findings } — the merged, ranked list (priority then confidence), where each kept finding carries { issue, impact, recommendation, priority, lenses, confidence, justification }.

3. Present and Confirm

After the workflow returns, present the kept findings to the user (group by priority: critical > important > nice-to-have). If findings is empty, tell the user the plan passed review and stop.

Use AskUserQuestion:

Question: "Based on these [N] issues, how would you like to proceed?"
Header: "Update Plan"
Options:
  - "Fix all" - Update plan to address all issues
  - "Select items" - Choose specific issues to address
  - "Skip" - Keep plan as-is

4. Update Plan (Only After Confirmation)

If user confirms:

  • Make targeted edits
  • Keep changes minimal
  • Preserve original structure

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.

The English summary is being prepared. Showing the original description.

SDiamante13/dotfiles11Updated Oct 2, 2026

Rebuilds session context after a clear by reading git history, plans, and beads state. Triggers on 'catch up'.

The English summary is being prepared. Showing the original description.

SDiamante13/dotfiles11Updated Oct 2, 2026

Generate a bullet-point summary of branch changes for a PR description. Triggers on "changelog", "pr summary", "summarize branch", or "/changelog".

The English summary is being prepared. Showing the original description.

SDiamante13/dotfiles11Updated Oct 2, 2026

commit

Free

Commit the current changes with quality checks, status review, staged diff inspection, appropriate tests, and a clear commit message. Use when the user says "commit", "c", asks to make a commit, or wants changes saved in git.

The English summary is being prepared. Showing the original description.

SDiamante13/dotfiles11Updated Oct 2, 2026

Refactor code to remove dead code, useless comments, unused variables, and unused imports. Use when asked to declutter a class, clean up noisy code, remove cruft, or run `/refactor:declutter`.

The English summary is being prepared. Showing the original description.

SDiamante13/dotfiles11Updated Oct 2, 2026

dry

Free

Refactor duplicated code by extracting helpers, shared abstractions, or clearer reusable structure. Use when asked to apply DRY, remove duplication, consolidate repeated logic, or run `/refactor:dry`.

The English summary is being prepared. Showing the original description.

SDiamante13/dotfiles11Updated Oct 2, 2026

See all skills by SDiamante13

Report a problem with this skill