super-product-owner — Claude Code skill: what it does, how to install
super-product-owner turns vague product ideas into a one-page brief, an interactive prototype, and concrete acceptance criteria. It narrows project scope before any production code gets written.
no license ★ 406 updated 2026-11-01
npx skills add syahiidkamil/Software-Engineer-AI-Agent-Atlas --skill super-product-owner -a claude-codeWhat super-product-owner does
The skill initiates a focused discovery dialogue. It poses exactly one inquiry per turn, targeting the most critical unknowns and rejecting fuzzy definitions. Instead of drowning in endless surveys, the interview wraps up in 5 to 12 prompts. The outcome is /docs/brief.md, capturing the primary user pain, the riskiest bet, and clear exclusions that prevent scope creep.
Once the brief is ready, it hands off specs to /brainstorm:prototype or /brainstorm:wireframe. The resulting interactive mockup undergoes a taste check to confirm whether the core interaction genuinely lands or feels mediocre. Finally, it outlines visible verification steps ready for execution by manual-qa-haiku.
This skill will not write implementation code. Its entire focus is pruning unnecessary requests before development begins. If discovery proves the underlying problem is negligible, the agent advises dropping the idea entirely.
What it helps with
- scope an MVP without creating bloated backlogs
- conduct a focused discovery interview one question at a time
- document risky assumptions and deliberate cuts in a single page
- evaluate an interactive wireframe before engineering starts
- define observable acceptance criteria instead of user stories
Example prompts for the agent
- Interview me with super-product-owner to define the core value of our new analytics tool
- Draft a one-page brief in docs/brief.md for our recurring billing feature
- Review the current interactive prototype and generate the acceptance check list
How to install super-product-owner
1Skills CLI (Vercel)
npx skills add syahiidkamil/Software-Engineer-AI-Agent-Atlas --skill super-product-owner -a claude-code2Manually: put the files in place
~/.claude/skills/super-product-owner/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/super-product-owner/
1Skills CLI (Vercel)
npx skills add syahiidkamil/Software-Engineer-AI-Agent-Atlas --skill super-product-owner -a codex2Manually: put the files in place
.agents/skills/super-product-owner/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add syahiidkamil/Software-Engineer-AI-Agent-Atlas --skill super-product-owner -a cursor2Manually: put the files in place
.cursor/skills/super-product-owner/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install syahiidkamil/Software-Engineer-AI-Agent-Atlas .claude/skills/super-product-owner2Manually: put the files in place
.github/skills/super-product-owner/SKILL.md1Skills CLI (Vercel)
npx skills add syahiidkamil/Software-Engineer-AI-Agent-Atlas --skill super-product-owner -a gemini-cli2Platform CLI
gemini skills install https://github.com/syahiidkamil/Software-Engineer-AI-Agent-Atlas --path .claude/skills/super-product-ownerManually: put the files in place
.devin/skills/super-product-owner/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about super-product-owner
How do I install super-product-owner in Claude Code?
Run npx skills add syahiidkamil/Software-Engineer-AI-Agent-Atlas --skill super-product-owner -a claude-code. The skill file installs to ~/.claude/skills/super-product-owner/SKILL.md.
Does super-product-owner write application source code?
No, it strictly handles product discovery, briefs, and acceptance definitions. It leaves coding and UI implementation to separate tools.
How does the skill handle missing information during discovery?
It presents clear multiple-choice options or explicitly documents the unknown as a project risk in the brief. Leaving a verified gap is preferred over fabricating assumptions.