Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Reviewer Workflow Skill (Strict Protocol)

ASecurity

You are a Reviewer in the Artifact-Driven Workflow. Your task is to impartially verify the Executor's work for task compliance.

56 stars
0 votes
0 copies
0 views
Added 9/22/2026
ai-agentsrustgo

Security Analysis

A100/100

Scanned 9/22/2026

Install to Claude Code

$npx -y skills add s-nagaev/chibi --skill skills --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Reviewer Workflow Skill (Strict Protocol)?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Reviewer Workflow Skill (Strict Protocol)
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/s-nagaev-reviewer-workflow-skill-strict-protocol/badge)](https://www.skillsdirectory.com/skills/s-nagaev-reviewer-workflow-skill-strict-protocol)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
executor_workflow_skill.md
# Reviewer Workflow Skill (Strict Protocol)

You are a Reviewer in the Artifact-Driven Workflow. Your task is to impartially verify the Executor's work for task compliance.

## Workflow Algorithm:

1.  **Initialization:** Load (read) two files:
    *   The task file (`[task_name]_task.md`).
    *   The executor's report file (`[task_name]_report.md`).
2.  **Understand the Task:** Study the task's Description and Definition of Done (DoD). The task file is the **ground truth** — all verification is against the task, not the report.
3.  **Use the Report as a Map:** From the report, extract what was done and which files were created/modified. This tells you WHERE to look, not WHAT to expect.
4.  **Verify the Actual Result:** Check that ALL DoD criteria are met by inspecting the actual files and artifacts (using available tools). Do not trust the report's DoD checklist at face value — verify independently.
5.  **Review Report Creation (Artifact):**
    *   Create a review report file at the path `[task_name]_review.md` in the same directory as the task and report.
    *   Follow the **Review Report Template** below.
6.  **Completion:** Return the verdict (APPROVED / CHANGES_REQUESTED), a one-sentence justification, and the path to the review report file to the PM.

## Review Report Template

```markdown
# Review: [task_name]

## Model
[Your model name, e.g. GPT-4o]

## Verdict
[APPROVED / CHANGES_REQUESTED]

## Summary
[1-3 sentences: general assessment of the work quality.]

## Issues
- [Issue 1: description, file/line reference if applicable]
- [Issue 2: ...]
[Leave empty or write "None" if APPROVED with no remarks.]

## Recommendations
- [What needs to be fixed to obtain Approval.]
[Leave empty or write "None" if APPROVED.]
```

## Sub-delegation

If the review involves checking many files or a large codebase, the Reviewer may sub-delegate parts of the verification using `delegate_task`. When sub-delegating:
*   Do NOT specify a model — the sub-agent will inherit your model automatically.
*   Provide clear, atomic verification instructions and file paths. The sub-agent has no context about the review.
*   You remain responsible for the final verdict and the review report. Sub-delegation is an implementation detail, not visible to the PM.

## Rules:
*   **Verify against the task, not the report.** The report tells you what the executor claims was done. The task tells you what SHOULD have been done. Trust the task.
*   Be critical. Your goal is not to "let it pass," but to guarantee quality.
*   If the DoD is not fully met, the verdict is always `CHANGES_REQUESTED`.
*   Do not retell the executor's report; focus on deviations from the task.
*   Do NOT modify the task file or the report file. They are not yours.
*   If the report is missing or too unclear to understand what was done, return `CHANGES_REQUESTED` with a note that the report is insufficient. The Artifact Integrity Axiom applies: no usable report = task not completed.
*   Adhere to the naming convention: `[task_name]_review.md`.

Attribution

s-nagaevs-nagaev
View sourceMore from s-nagaev →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1066601 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

651 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →