michael-denyer/black-box-unlock/review-change

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Review the current branch, index, or working tree with calibrated repository evidence from bbu. Use before opening a PR or when the user asks what to check in a change.

Install

npx vskill@latest install michael-denyer/black-box-unlock/review-change

SKILL.md

Review the change that is about to ship:

  1. Run bbu review-change --base origin/main, or call the review_change MCP

tool with mode="base" and base_ref="origin/main" when the black-box-unlock server is connected. The tool's default mode is working_tree, which skips branch commits.

  • If origin/main is not the intended upstream, use the base named by the

user.

  • For index-only review, run bbu review-change --staged.
  • For unstaged and untracked work only, run `bbu review-change

--working-tree`.

  • Use --profile NAME when .bbu.toml defines a profile for this review.
  • Use --include-ci only when failed GitHub workflow evidence is useful.
  1. Report the returned actions in order. Preserve the exact paths, raw

revision counts, directional coupling rate, and ratio_lower_bound, the 95% lower bound of the symmetric ratio.

  1. Treat an empty action list as a valid clean result. Do not invent extra

recommendations or turn the evidence into a composite risk score. If omitted_actions is non-zero, say how many candidates the cap dropped.

  1. Mention whether CI was disabled, available, partial, or unavailable. When an

inspect_ci_failures action is present, link the returned run and say that its changed paths are implicated, not proven causal.

  1. Repeat any warnings the result carries, such as a shallow clone.

The command applies the selected profile, project path-role rules, support floor, deterministic ordering, and three-action cap. History before a rename counts toward the current path.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/michael-denyer/black-box-unlock/review-change/badge)](https://verified-skill.com/skills/michael-denyer/black-box-unlock/review-change)

Something wrong with this skill or its page? Tell us.