# Code Review

> Protocol for reviewing code and providing structured verdicts

- **Type:** Skill
- **Install:** `agentstack add skill-omnigentx-jarvis-code-review`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [omnigentx](https://agentstack.voostack.com/s/omnigentx)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [omnigentx](https://github.com/omnigentx)
- **Source:** https://github.com/omnigentx/jarvis/tree/main/backend/.fast-agent/skills/code-review

## Install

```sh
agentstack add skill-omnigentx-jarvis-code-review
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Code Review Skill

You are performing a code review. Follow this protocol for consistent, actionable feedback.

## Review Process

### Step 1: Understand the Context
- Read the project brief and any related documentation in the workspace
- Understand what the author was trying to accomplish
- Check `team_roster.json` to know who wrote what

### Step 2: Review the Code
- Read all relevant files in the workspace
- Check for:
  - **Correctness** — Does the code do what it's supposed to?
  - **Quality** — Is it readable, maintainable, well-structured?
  - **Edge cases** — Are there unhandled scenarios?
  - **Security** — Any obvious security issues?
  - **Performance** — Any obvious performance problems?

### Step 3: Write Your Review
Write your review to: `reviews/_review.md`

Include:
1. **Summary** — Overall assessment
2. **Issues found** — Specific problems with file/line references
3. **Suggestions** — Improvements that aren't blockers
4. **Verdict** — Your final decision

### Step 4: Declare Verdict

**Always end your review with one of these exact strings:**

✅ When code is acceptable:
```
[DECISION] VERDICT: PASS — 
```

❌ When code needs fixes:
```
[DECISION] VERDICT: FAIL — 
```

## Communication

After writing your review:
- Use `send_email(to="", body="...", my_name="")` to notify the author
- Include a summary of key issues if FAIL
- Be constructive — suggest fixes, don't just point out problems

## As a Code Author (receiving review)

When you receive a FAIL verdict:
1. Read the review feedback carefully
2. Fix the identified issues in your workspace files
3. Message the reviewer: `send_email(to="", body="Fixes applied, ready for re-review", my_name="")`

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [omnigentx](https://github.com/omnigentx)
- **Source:** [omnigentx/jarvis](https://github.com/omnigentx/jarvis)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-omnigentx-jarvis-code-review
- Seller: https://agentstack.voostack.com/s/omnigentx
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.