# Review

> |

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

## Install

```sh
agentstack add skill-nathanfaucett-agents-review
```

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

## About

## Summary

Use this skill for static content quality reviews that produce clear findings and practical improvements.

## When to use
- Reviewing documents for clarity, correctness, and completeness.
- Reviewing code for duplication, maintainability, and readability risks.
- Reviewing datasets for obvious consistency or redundancy issues.
- Providing a second opinion before publishing or merging content.

## When not to use
- Runtime debugging or execution failures that require running systems or logs.
- Security audits or compliance checks that need specialized security workflows.
- Tasks requiring direct implementation rather than analysis.
- Reviews that depend on unavailable external business context.

## Inputs
- Content to review (document, code, or data).
- Requested focus (for example: clarity, duplication, completeness, consistency).
- Optional context: style guide, acceptance criteria, constraints, or target audience.

## Outputs
- Structured review report with:
  1. Key findings (prioritized)
  2. Duplications
  3. Clarity/completeness gaps
  4. Accuracy/consistency issues
  5. Actionable improvements
  6. Open questions (if context is missing)

## Procedure
1. Clarify scope only if review focus is ambiguous.
2. Inspect content for duplication, ambiguity, inconsistency, and omission.
3. Apply provided standards or default community conventions.
4. Produce a concise, prioritized report with actionable fixes.
5. Separate confirmed findings from context-dependent questions.

## Best practices and constraints
- Keep findings specific and evidence-based.
- Distinguish objective issues from style preferences.
- Prefer actionable recommendations over generic commentary.
- Do not claim runtime behavior; this skill is static analysis only.

## Gotchas and edge cases
- Missing context can make valid patterns appear inconsistent; surface as open questions.
- Very small snippets can hide dependencies; avoid overconfident conclusions.
- Domain-specific correctness may require expert validation beyond this review.

## Examples
- "Review this requirements doc for missing acceptance criteria and ambiguity."
- "Review this module for duplication and maintainability risks."
- "Review this dataset description for consistency and clarity."

## Acceptance criteria
- Report is structured, prioritized, and actionable.
- Findings clearly separate confirmed issues from open questions.
- Recommendations are tied to observed content, not speculation.

## Notes for agents
- Ask at most one clarifying question when scope is unclear.
- Cite exact sections or lines when available.
- Keep output concise and decision-oriented.

## Limitations
- Does not execute code or validate runtime behavior.
- May miss highly domain-specific issues without additional context.
- Feedback quality depends on the completeness of provided inputs.

## Source & license

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

- **Author:** [nathanfaucett](https://github.com/nathanfaucett)
- **Source:** [nathanfaucett/agents](https://github.com/nathanfaucett/agents)
- **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-nathanfaucett-agents-review
- Seller: https://agentstack.voostack.com/s/nathanfaucett
- 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%.