bug-reviewer
CLAUDE.md0 starsChanged 6 months agoMITView on GitHub
The bug-fix workflow lives in skills, not in this file. This file is a thin pointer — see .claude/skills/ for everything else. When the user asks to fix bugs, triage new bugs, run a bug session, or ship bug fixes — invoke the bug-triage-session skill. It orchestrates the full flow and pulls in the domain skills as needed. One-time human setup (Node, pnpm, Azure CLI, Playwright, Python, Figma MCP) lives in README.md. If your codebase has design-system conventions (color tokens,…
- Reads credentials
In this repository8 other files it gives its agents.
- ado-bug-lifecycleSkill
- ado-bug-querySkill
- ado-pr-createSkill
- bug-captureSkill
- bug-reviewer-runSkill
- bug-triage-sessionSkill
- storyblok-bugfix-storySkill
- storyblok-component-pushSkill
The file
We split this CLAUDE.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 9
When the user asks to fix bugs, triage new bugs, run a bug session, or ship bug fixes — invoke the bug-triage-session skill.
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Bug Reviewer — Claude Playbook
The bug-fix workflow lives in skills, not in this file. This file is a thin pointer — see .claude/skills/ for everything else.
Entry point
When the user asks to fix bugs, triage new bugs, run a bug session, or ship bug fixes — invoke the bug-triage-session skill. It orchestrates the full flow and pulls in the domain skills as needed.
Available skills (in .claude/skills/)
| Skill | When it loads |
bug-triage-session | Top-level orchestrator for an end-to-end bug session |
ado-bug-query | Listing new bugs from the configured ADO board |
ado-bug-lifecycle | Picking up / abandoning bugs (state + assignee) |
ado-pr-create | Creating a draft PR with reviewers + work items |
storyblok-bugfix-story | Creating, finding, or publishing a bugfix story |
storyblok-component-push | Pushing a new bugfix_* component schema |
bug-capture | Running capture.js for before/after screenshots |
bug-reviewer-run | Launching the local reviewer UI |
One-time human setup (Node, pnpm, Azure CLI, Playwright, Python, Figma MCP) lives in README.md.
Adding a design-reference skill
If your codebase has design-system conventions (color tokens, spacing variants, common component patterns, gradient names, max-width policies), add a domain-reference skill at .claude/skills/<your-design>-reference/SKILL.md. The skill should load on triggers like "backdrop", "gradient", "max-width", "spacing token" — whatever the agent is likely to encounter when fixing visual bugs in your codebase — and supply reference data the agent would otherwise have to grep for. Keep it scoped to reference data; commands and rules belong in domain skills.
Cross-domain hard rules
These apply across all skills — keep them in mind whenever working in this repo:
- No AI references in committed files, commit messages, or PR descriptions.
- The dev server must be running at
$PREVIEW_BASE_URL(defaulthttps://localhost:3000) before any screenshot capture. - Never touch CMS stories or components outside the bug-fix scope (e.g. the
pages/dev/bugs/folder andbugfix_*components for Storyblok). Treat the rest of the CMS as read-only. - Source
.envbefore starting:set -a && source .env && set +a. The skills assume$AZURE_ASSIGNEE_EMAIL,$STORYBLOK_MANAGEMENT_TOKEN,$DRAFT_SECRET_TOKEN, and$TARGET_REPO_PATHare available.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.