sandgardenhq/sgai/requesting-code-review
Claude Code and agent skill by sandgardenhq/sgai
- community
Checking that the SKILL.md file is still in the repository.
Use when completing tasks, implementing major features, or before merging to verify work meets requirements - dispatches bounded reviewer subagents with TASK_PACKET and REVIEW_STATUS envelopes before proceeding
Install
npx vskill@latest install sandgardenhq/sgai/requesting-code-reviewSKILL.md
Requesting Code Review
Dispatch the appropriate reviewer subagent with a bounded TASK_PACKET to catch issues before they cascade.
Core Principle
Review early, review often, and make every reviewer output actionable through REVIEW_STATUS.
When to Request Review
Mandatory:
- After completing a major feature
- Before marking a GOAL item complete
- Before merge or release
- After resolving non-trivial reviewer findings
Optional but valuable:
- When stuck and needing a fresh read-only perspective
- Before refactoring to establish baseline concerns
- After fixing a complex bug
How to Request
- Identify the exact review scope:
- Changed paths or
jj diffscope - Requirement or GOAL item being verified
- Developer
STATUSenvelope evidence - Verification commands already run
- Screenshots, deployment URLs, or artifacts when relevant
- Dispatch the domain reviewer subagent:
go-reviewerfor Go codereact-reviewerfor React/TypeScript UI codehtmx-picocss-reviewerfor HTMX/PicoCSS UIshell-script-reviewerfor shell scriptswebmaster-reviewerfor website/UI/content/accessibility/SEO readiness
- Use a bounded packet:
Load and follow subagent-delegation-protocol skill, then send the reviewer its Task Packet. Set expected edit boundaries to read-only review and set the return envelope to the Review Envelope from subagent-return-protocol skill.
- Act on feedback:
PASS: reviewer found no required fixes in scopeNEEDS_WORK: route findings back to the developer; every finding is blockingNEEDS_CONTEXT: provide only the missing path, evidence item, decision, or question requestedBLOCKED: record the blocker and route it to the owner
Example
Integration With Workflows
Executing plans:
- Review after each independent work unit or batch of non-overlapping units
- Loop
NEEDS_WORKfindings back to the developer - Re-review until
PASSor an explicit blocker remains
Ad-hoc development:
- Review before merge or completion
- Review when stuck
Red Flags
Never:
- Skip review because "it's simple"
- Ignore
NEEDS_WORK - Proceed with unresolved reviewer findings
- Ask a reviewer to rediscover the whole repository when paths are known
- Send review requests without bounded paths or evidence
If reviewer is wrong:
- Push back with technical reasoning
- Show code/tests that prove it works
- Request clarification with a focused packet
Reviewer Output
Reviewers must return this envelope exactly:
Load and follow subagent-return-protocol skill, then return its Review Envelope exactly.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingcmd/sgai/skel/.sgai/skills/requesting-code-review/SKILL.md
Popularity
- Repository stars
- 124
- Repository forks
- 7
- 7-day trend score
- 24
Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 27 Mar 2026, 7 months ago.
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
[](https://verified-skill.com/skills/sandgardenhq/sgai/requesting-code-review)