From devflow
Use when creating or reviewing pull requests — PR description, review checklist, merge readiness
npx claudepluginhub nexuz-sys/devflow --plugin devflowThis skill uses the workspace's default tool permissions.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
Standards for creating high-quality pull requests and performing thorough reviews.
Announce at start: "I'm using the devflow:pr-review skill."
## Summary
[1-3 bullet points explaining WHAT and WHY]
## Changes
- [Key files/modules changed and why]
## Testing
- [ ] Unit tests added/updated
- [ ] Integration tests pass
- [ ] Manual testing done (describe scenarios)
## Screenshots
[If UI changes — before/after]
## Breaking Changes
[Any breaking changes and migration path, or "None"]
## Related
- Closes #123
- Depends on #456
- Design spec: docs/specs/YYYY-MM-DD-feature.md
**BLOCK:** [Must fix before merge]
Line 42: SQL injection via string concatenation — use parameterized query
**WARN:** [Should fix, not a blocker]
Line 78: This could be simplified with Array.filter()
**NOTE:** [Suggestion, take it or leave it]
Line 105: Consider extracting this to a helper if it's used elsewhere
skill({ action: "getContent", skill: "pr-review" })
agent({ action: "orchestrate", agents: ["code-reviewer"], task: "pr-review" })
Read .context/skills/pr-review/SKILL.md and .context/agents/code-reviewer.md.
For detailed code review, invoke superpowers:requesting-code-review (as reviewer) or superpowers:receiving-code-review (processing feedback).
| Pattern | Problem |
|---|---|
| "LGTM" without reading | Rubber-stamp reviews catch nothing |
| Reviewing only the happy path | Bugs live in edge cases and error paths |
| Style nitpicks over substance | Focus on correctness and maintainability first |
| Giant PRs | Impossible to review thoroughly — split them |
| No description | Reviewer needs context to give useful feedback |