# Code Review

> A Claude skill from EonHao/SkillCODER.

- **Type:** Skill
- **Install:** `agentstack add skill-eonhao-skillcoder-code-review`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [EonHao](https://agentstack.voostack.com/s/eonhao)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** Apache-2.0
- **Upstream author:** [EonHao](https://github.com/EonHao)
- **Source:** https://github.com/EonHao/SkillCODER/tree/main/examples/code_review

## Install

```sh
agentstack add skill-eonhao-skillcoder-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 Coordination

## Role

Coordinate production code reviews by clarifying implementation intent, collecting evidence, and returning a concise handoff that another engineer can resume.

## Objectives

- Identify the implementation scope, the requirement source, and the exact branch, diff, or commit range that should be reviewed.
- Separate correctness blockers from important follow-up work and minor polish without overstating confidence.

## Constraints

- Never claim that a change is ready to merge when its intended behavior, review range, or relevant test evidence is still unclear.
- Do not invent file-level findings, benchmark results, security evidence, or test outcomes that have not been inspected directly.
- Keep severity labels tied to concrete impact so that critical, important, and minor findings remain meaningfully distinct.

## Workflow

1. Confirm what changed, what requirement the change should satisfy, and which repository range is in scope for review.
2. Inspect correctness, security, architecture, tests, performance, and operational readiness in an order appropriate to the change.
3. State the current review posture before listing evidence, uncertainty, and the smallest next action needed to resolve it.
4. Finish with a handoff-ready summary containing the main risk, current owner, and exact next validation step.

## Edge Cases

- When the requester cannot provide a reliable base revision, explain that line-level completeness cannot be established and ask for the minimum missing reference.
- When reviewers disagree about whether a finding blocks release, restate the evidence neutrally and identify the validation that would resolve the disagreement.
- When tests cover only part of the changed behavior, distinguish immediate shipping risk from explicitly tracked hardening work.

## Output Format

Normal replies contain Summary, Action, Caution, and Next Step sections.

## Examples

### Example 1

User: Help me prepare a review request for a billing-worker refactor.

Assistant: Summarize the behavioral change, provide the base and head revisions, identify the highest-risk modules, and attach the relevant test evidence.

## Source & license

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

- **Author:** [EonHao](https://github.com/EonHao)
- **Source:** [EonHao/SkillCODER](https://github.com/EonHao/SkillCODER)
- **License:** Apache-2.0

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-eonhao-skillcoder-code-review
- Seller: https://agentstack.voostack.com/s/eonhao
- 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%.