tsaol/awesome-claude/product-review
Claude Code and agent skill by tsaol/awesome-claude
- community
Checking that the SKILL.md file is still in the repository.
Product review from a PM perspective. Uses multi-agent pipeline to analyze any project/codebase for product discovery, goal alignment, UX/DX audit, competitive analysis, and roadmap planning.
Install
npx vskill@latest install tsaol/awesome-claude/product-reviewSKILL.md
Product Review - 多 Agent 产品经理评审
以产品经理视角全面评审任意项目/代码库,输出结构化评审报告和改进路线图。
Usage
Options:
| Option | Description | Default | ||
|---|---|---|---|---|
--quick | Quick mode (skip competitive analysis, simplified roadmap) | false | ||
--focus=<area> | Focus area: ux / dx / growth / tech | all | ||
| `--lang=<zh\ | en>` | Output language | zh |
Examples:
Output Structure
All outputs are generated in the target project's root directory:
Pipeline
Full Mode (5 Agents)
Quick Mode (3 Agents, skip Agent 4)
Execution Steps
Step 0: Setup
Step 1: Phase 0 - Project Snapshot
Non-LLM step. Collect project metadata using bash commands and write to .product-review/raw/snapshot.md.
Collect the following:
Write to .product-review/raw/snapshot.md:
Log:
Step 2: Agent 1 - Product Discovery
Run the pr-discovery agent.
Input: Read .product-review/raw/snapshot.md Output: Agent writes to .product-review/analysis/discovery.md
Step 3: Agent 2 + Agent 3 (Parallel)
Run pr-goal-alignment and pr-ux-dx-audit in parallel using two Task tool calls in a single message.
Agent 2 Input: .product-review/raw/snapshot.md + .product-review/analysis/discovery.md Agent 2 Output: .product-review/analysis/goal-alignment.md
Agent 3 Input: .product-review/raw/snapshot.md + .product-review/analysis/discovery.md Agent 3 Output: .product-review/analysis/ux-dx-audit.md
Step 4: Agent 4 - Competitive Analysis (Full Mode Only)
Skip this step if `--quick` is set.
Run the pr-competitive-analysis agent.
Input: .product-review/raw/snapshot.md + .product-review/analysis/discovery.md Output: .product-review/analysis/competitive-analysis.md
Step 5: Agent 5 - Roadmap Planner (Final Report)
Run the pr-roadmap-planner agent.
Input: All files in .product-review/analysis/ + .product-review/raw/snapshot.md Output: .product-review/output/pm-review.md
Step 6: Summary
After all agents complete, print:
Focus Mode Behavior
When --focus=<area> is set, agents should weight their analysis accordingly:
| Focus | Agent Emphasis | |
|---|---|---|
ux | Agent 3 deep-dives into user experience, UI flows, error messages | |
dx | Agent 3 deep-dives into developer experience, API design, docs, SDK | |
growth | Agent 1 emphasizes growth potential, Agent 4 emphasizes market opportunity | |
tech | Agent 2 emphasizes technical implementation quality, architecture |
Pass the focus area to each agent via a note at the top of the snapshot:
Language Behavior
When --lang=en is set, all agents output in English. Default is Chinese (zh).
Pass the language preference to each agent via a note at the top of the snapshot:
Log Format
All pipeline logs are appended to .product-review/process/pipeline.log:
Status symbols:
▶started✓done⚠warning✗failed
Final Report Template
The final report (output/pm-review.md) follows this template:
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/product-review/SKILL.md
Popularity
- Repository stars
- 62
- Repository forks
- 4
- 7-day trend score
- 21
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
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
[](https://verified-skill.com/skills/tsaol/awesome-claude/product-review)