"AI 产品经理教练。通过引导式对话帮助 PM 完成 AI 产品设计：从痛点分析到 PRD 输出。不替代 PM 决策，而是引导 PM 思考，在关键节点让 PM 做出选择。触发词：AI 产品、产品设计、PRD、能力边界、置信度、幻觉。" metadata:
v1.0.5AI 产品经理教练。通过引导式对话帮助 PM 完成 AI 产品设计：从痛点分析到 PRD 输出。不替代 PM 决策，而是引导 PM 思考，在关键节点让 PM 做出选择。触发词：AI 产品、产品设计、PRD、能力边界、置信度、幻觉。
⭐ 0· 66·0 current·0 all-time
bySocialite UCL LJH@lijinhongucl-pixel
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description, the SKILL.md, ARCHITECTURE.md, README and the five internal skill manifests all describe an interactive coaching workflow (pain check, boundary, confidence, hallucination, PRD). No unrelated credentials, binaries, or config paths are requested.
Instruction Scope
Runtime instructions are dialogue-focused and describe calling internal sub-skills with structured JSON inputs/outputs. The SKILL.md does not instruct the agent to read files, access unrelated environment variables, or send data to external endpoints. It explicitly keeps internal skills internal and pauses for user decisions.
Install Mechanism
No install spec or code is present (instruction-only Sub Agent). Nothing will be downloaded or written to disk by an installer as part of the skill package.
Credentials
The skill declares no required environment variables, credentials, or config paths. The data used is conversational (product design inputs) which is appropriate for a coaching skill.
Persistence & Privilege
always:false and no special persistence or system-wide config changes are requested. The skill is allowed to be invoked autonomously (platform default), but it does not request elevated or permanent privileges.
Scan Findings in Context
[no_regex_findings] expected: Scanner had no code files to analyze (instruction-only skill). Absence of findings is expected for an instruction-only / subagent package.
Assessment
This skill appears coherent and safe in architecture: it only guides a PM through structured questions and calls internal sub-skills to produce a PRD draft. Before installing, consider: (1) conversational inputs may include confidential product details — avoid pasting sensitive secrets or proprietary data unless you trust where the agent runs and how data is stored/retained; (2) confirm the runtime/execution environment (local vs. remote) and platform privacy/retention policies so you know whether conversations are logged or sent off-platform; (3) the skill intentionally delegates decisions to the user — always review and approve any PRD or policy it generates; (4) if you plan to integrate this with other systems, verify any connectors or grants required at integration time are proportionate. If you want additional assurance, ask the author for a data handling/privacy statement and where the sub-agent execution occurs.Like a lobster shell, security has layers — review code before you run it.
latestvk972e7j5kdtzg61k42g0r3r7eh84ve0g
License
MIT-0
Free to use, modify, and redistribute. No attribution required.