
Review Pr
Runs a multi-agent PR review with 6-7 specialized reviewers analyzing code quality, security, testing, architecture, and performance in parallel. You pass a PR number or branch, it fetches the diff via gh CLI, then spawns agents that each focus on one dimension and produce conventional comments (praise, issue, suggestion, nitpick) plus an approve or request-changes verdict. The workflow is surprisingly thorough: it probes for the memory MCP server to recall patterns from past reviews, asks what kind of review you want (full, security-focused, quick), and on Claude Code 2.1.120+ can defer to the native ultrareview subcommand. Good for teams that want consistent, structured reviews without the overhead of setting up separate CI linters. The task tracking is verbose but keeps long reviews organized.
- skills.sh installs
- 217
- GitHub stars
- 301
- last updated
- Aug 21, 2026
Install
$ npx -y skills add yonatangross/orchestkit --skill review-pr --agent claude-codeInstalls into .claude/skills of the current project.
Files
Review PR
Host-neutral workflow. Invoke by skill name (review-pr). Claude Code slash routing, YAML hook loaders, and .claude/chain live in references/claude-code.md.
Deep code review using 6-7 parallel specialized agents.
Quick Start
review-pr 123
review-pr feature-branch
Opus 5.5: Parallel agents use native adaptive thinking for deeper analysis. Complexity-aware routing matches agent model to review difficulty.
Argument Resolution
A pronoun target (this, them, the above) is read back from the conversation, announced in one line, and used: Read("../../shared/rules/target-resolution.md"), inline half.
Resolve the target with the script first, then review exactly that target (#3892). Read the mode and opt-in flags (--rules, --standards, --post, --post-verdict) from the arguments yourself and drop them, then run the script as one plain command (the post gate allows only simple read-only commands, so no assignment, loop or $(...)):
bash ${CLAUDE_SKILL_DIR}/scripts/resolve-target.sh <the arguments without the four flags> # prints one JSON object: TARGET
Sponsors
Details
- Repository
- github.com/yonatangross/orchestkit
- Categories
- Testing & QA, Security, Git & Pull Requests, Code Review & Quality, Debugging, Automation & Workflows
- First seen
- Jun 3, 2026
More from yonatangross/orchestkit
All 61- Llm Integration216
- I18n Date Patterns215
- Task Dependency Patterns214
- Api Design211
- Assess211
- Analytics210
- Commit210
- Skill Evolution210
- Verify210
- Create Pr209
- Audit Full206
- Implement206
- Configure196
- Demo Producer195
- Async Jobs194
- Golden Dataset193
- Product Frameworks193
- Browser Tools191
- Mcp Patterns190
- Feedback180
- Scope Appropriate Architecture179
- Testing Patterns179
- Help166
- Checkpoint Resume156