Product Manager Skills

PM skill for Claude Code, Codex, Cursor, and Windsurf. Diagnoses SaaS metrics, critiques PRDs, plans roadmaps, runs discovery, coaches PM career transitions,...

This skill appears coherent and low-risk: it is Markdown-first, asks for no secrets, and explicitly disallows automatic execution of local helper scripts. Before installing, skim SKILL.md and the small bin scripts (bin/update-check, bin/validate-release) yourself — the README says they are manual helpers but you should verify their contents if you plan to run them. If you install via npx or clawhub, treat that like any third-party package: inspect package.json and any postinstall hooks. When using the skill, only supply metrics and reports you consent to share; the skill will ask you to pull data (e.g., from Mailchimp/Stripe) but does not itself request access tokens or automatically fetch them. Finally, remember the skill can be invoked autonomously by the agent (platform default) — that is normal here, but if you want stricter control, disable or restrict autonomous skill invocation at the agent/platform level.

Type: OpenClaw Skill Name: product-manager-skills Version: 0.5.4 The 'product-manager-skills' bundle is a comprehensive Product Management persona implementation consisting of markdown-based frameworks, templates, and workflow instructions. It demonstrates strong security hygiene by explicitly instructing the AI agent in 'SKILL.md' to avoid automatic execution of local helper scripts (such as 'bin/update-check'), ensuring that any maintenance actions require manual user approval. The content is entirely aligned with its stated purpose, lacks any high-risk behaviors like data exfiltration or unauthorized network access, and shows an evolutionary improvement in security practices in its 'CHANGELOG.md'.