From kitchen-sink
Streamlined code review workflow - gets SHAs and invokes superpowers:code-reviewer without permission prompts
npx claudepluginhub oalders/kitchen-sink --plugin kitchen-sinkThis skill uses the workspace's default tool permissions.
Wrapper around `superpowers:requesting-code-review` that avoids permission prompts by using information already in context.
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.
Creates isolated Git worktrees for feature branches with prioritized directory selection, gitignore safety checks, auto project setup for Node/Python/Rust/Go, and baseline verification.
Wrapper around superpowers:requesting-code-review that avoids permission prompts by using information already in context.
When user requests code review:
Option 1: Use context (BEST - no Bash needed)
git log output, or known commit referencegit commit output or git logOption 2: Run git commands separately (already allowed in most projects)
# Run these as SEPARATE Bash calls, not chained with &&
git rev-parse origin/main # or HEAD~1, or specific commit
git rev-parse HEAD
DON'T do this (triggers permission prompts):
# ❌ Compound command with variable assignment
BASE_SHA=$(git rev-parse origin/main) && HEAD_SHA=$(git rev-parse HEAD) && echo "BASE=$BASE_SHA HEAD=$HEAD_SHA"
Once you have the SHAs, invoke the code-reviewer subagent:
Task tool with subagent_type: superpowers:code-reviewer
Prompt template:
# Code Review Agent
You are reviewing code changes for production readiness.
**Your task:**
1. Review [WHAT_WAS_IMPLEMENTED]
2. Compare against [PLAN_OR_REQUIREMENTS]
3. Check code quality, architecture, testing
4. Categorize issues by severity
5. Assess production readiness
## What Was Implemented
[DESCRIPTION - Brief summary of what was built]
## Requirements/Plan
[PLAN_REFERENCE - Link to issue, plan doc, or inline description of requirements]
## Git Range to Review
**Base:** [BASE_SHA]
**Head:** [HEAD_SHA]
```bash
git diff --stat [BASE_SHA]..[HEAD_SHA]
git diff [BASE_SHA]..[HEAD_SHA]
[... rest of code-reviewer template from superpowers:requesting-code-review/code-reviewer.md]
## Posting Review Results
**After review completes, check if PR exists:**
```bash
gh pr list --head $(git branch --show-current) --json number,url
If PR exists:
gh pr commentIf no PR exists:
When the review finds issues, fix them automatically rather than just reporting:
Once the review finds no remaining issues:
/monitor-ci slash command exists in the current project's available skills/monitor-ci exists, invoke it to monitor CI status/monitor-ci does not exist, inform the user the review is completeUser: "request a code review using superpowers"
Step 1: Check context for SHAs
- Recent git commit showed: [fix-1065 d0e856b8]
- git log showed base: 4f940124
Step 2: Invoke code-reviewer
Task(superpowers:code-reviewer):
WHAT_WAS_IMPLEMENTED: Tag sorting fix with case-insensitive handling
PLAN_OR_REQUIREMENTS: Issue #1065 - sort tags by distance value
BASE_SHA: 4f940124
HEAD_SHA: d0e856b8
DESCRIPTION: Added parseDistanceTag() and case-insensitive regex
Step 3: Review found 1 major issue (missing nil check) and 2 minor issues
- Diff is 180 lines (under 400) → fix all issues
- Commit fixes: [fix-1065 a1b2c3d4]
Step 4: Re-run review with updated HEAD
Task(superpowers:code-reviewer):
BASE_SHA: 4f940124
HEAD_SHA: a1b2c3d4
... (same params, new HEAD)
Step 5: Review passes clean
Step 6: Check for PR
$ gh pr list --head fix-1065 --json number
[{"number": 123}]
Step 7: Post clean review to PR
$ gh pr comment 123 --body "[Complete review in markdown]"
✓ Review posted to PR #123
Step 8: Check for /monitor-ci
- /monitor-ci exists → invoke it
- (or: /monitor-ci not found → inform user review is complete)
This skill works because:
Bash(git rev-parse:*) is typically already allowed✅ No permission prompts during code review ✅ User can leave window while review runs ✅ Faster workflow - no blocking on permissions ✅ Uses information already in context when available ✅ Posts review to PR when one exists - keeps discussion centralized