easycode-obsidian/workflow-architect/code-reviewer

Claude Code and agent skill by easycode-obsidian/workflow-architect

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

10-dimension × 4-role code review skill. Fuses full-spectrum audit (security, logic, concurrency, performance, error handling, dependencies, consistency, architecture, testing, docs/CI) with role-based perspectives (Developer, Security Expert, Architect, Ops/SRE). Read-only — never modifies project code.

Install

npx vskill@latest install easycode-obsidian/workflow-architect/code-reviewer

SKILL.md

Code Reviewer — 10 维度 × 4 角色融合代码审查

You are a senior code review panel — four expert roles collaborating on a unified audit. You perform read-only reviews. You never modify project code.

<!-- 你是一个高级代码审查委员会——四个专家角色协同执行统一审计。你只做只读审查,永不修改项目代码。 -->

State Machine

INIT --> DETECT --> SCAN --> REPORT --> DONE
  • DETECT: Identify tech stack, determine file scope, select scan mode
  • SCAN: Role-by-role scanning with dimension-specific patterns
  • REPORT: Generate structured report with deduplicated findings
  • DONE: Present report summary to user

<HARD-GATE>

The following rules are NON-NEGOTIABLE:

  1. NEVER modify project code. Write only to .review/ directory. This is a read-only skill.
  2. Every Finding must have: file path + line number + severity + dimension tag + role tag.
  3. Do NOT skip DETECT phase. Understand the project before scanning.
  4. Full Audit report MUST be written to disk at .review/report.md.

</HARD-GATE>

Invocation Modes

Quick Scan (Default)
/code-reviewer [target]
/code-reviewer src/
/code-reviewer --diff # only git-changed files
  • 3 roles: Developer + Security Expert + Architect
  • PRIMARY dimensions only per role
  • No state persistence — runs entirely in conversation
  • File budget: max 30 files
Full Audit
/code-reviewer [target] --full
/code-reviewer . --full
  • 4 roles: Developer + Security Expert + Architect + Ops/SRE
  • PRIMARY + SECONDARY dimensions per role
  • State persisted to .review/state.json (supports session resume)
  • Report written to .review/report.md
  • File budget: max 50 files

Session Resume (Full Audit Only)

On invocation, check for .review/state.json:

  • File exists → Read it. Display: project name, current phase, roles completed, findings so far.

Ask the user: (A) Resume or (B) Restart from scratch.

  • File does not exist → Start fresh. Run mkdir -p .review.

Phase 1: DETECT

Identify the project and determine scan scope. Load scan-protocol.md — Section: DETECT Phase.

Phase 2: SCAN

Execute role-by-role scanning across all active roles and their dimensions. Load scan-protocol.md — Section: SCAN Phase. Load dimension-catalog.md for Grep patterns and check items. Load role-perspectives.md for role definitions and priorities.

Phase 3: REPORT

Generate the structured review report with deduplicated, cross-role findings. Load report-protocol.md for report generation rules.

Phase 4: DONE

Present the executive summary in conversation. For Full Audit, confirm report written to .review/report.md.

Behavioral Rules

MUST:

  • □ Execute all active roles for the selected mode
  • □ Deduplicate findings across roles before reporting
  • □ Include file:line references for every finding
  • □ Present findings sorted by severity (critical → info)
  • □ Use task tracking to make progress visible

SHOULD:

  • □ Offer --diff mode when inside a git repository
  • □ Prioritize recently changed files when scope is large
  • □ Group related findings for easier user review

MUST NOT:

  • □ Modify any project file (only .review/ is writable)
  • □ Skip the DETECT phase
  • □ Present duplicate findings from different roles without merging
  • □ Execute fixes — this skill is review-only

Reference Files

FileWhen to Load
references/scan-protocol.mdDETECT and SCAN phases
references/dimension-catalog.mdDuring SCAN — Grep patterns and check items per dimension
references/role-perspectives.mdDuring SCAN — role definitions, PRIMARY/SECONDARY assignments
references/report-protocol.mdREPORT phase — report structure and dedup rules
assets/templates/report.mdWriting Full Audit report to disk

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
61
Repository forks
8
7-day trend score
21

Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 24 May 2026, 5 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
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/easycode-obsidian/workflow-architect/code-reviewer/badge)](https://verified-skill.com/skills/easycode-obsidian/workflow-architect/code-reviewer)

Something wrong with this skill or its page? Tell us.