{
  "markdown": "<p align=\"center\"><img src=\"logo.png\" width=\"120\" alt=\"logo\" /></p>\n\n# /superpm\n\n**English** | [한국어](README.ko.md)\n\n**One PM request in, a verified artifact out - the smallest useful artifact, checked by an\nindependent critic against the framework and the real decision.**\n\nA single agent skill that folds the whole PM toolbox into one workflow: it captures intent from\nyour request, **diverges by PM domain** into the right framework, produces the smallest useful\nartifact, and runs an independent critic over it before delivery - then stops.\n\nNo extra install: clone the repo, symlink it into your skills directory, then `/superpm <ask>`.\nLanding page: **[cskwork.github.io/superpm-skill](https://cskwork.github.io/superpm-skill/)**.\n\n## What it is\n\n`phuryn/pm-skills` is a marketplace of 68 PM skills across 9 plugins, each a separate file you\nload by name. `superpm` collapses that catalog into **one skill, one repo**. You do not pick a\nskill; you state the PM ask and the workflow routes it:\n\n```\n/superpm write a PRD for in-app referral\n/superpm what's our TAM/SAM/SOM for SMB payroll in Korea\n/superpm build an opportunity solution tree for activation\n/superpm draft a pricing strategy for the pro tier\n/superpm read this A/B test and tell me ship or kill\n/superpm a GTM plan for the EU launch\n/superpm review my PM resume against this job description\n```\n\n## How it works\n\nThe same loop runs for every domain; only the framework that loads changes.\n\n1. **Capture** - classify the ask into a PM domain + a target artifact. If genuinely ambiguous,\n   a <=5 question interview gate fires; anything answerable from your docs/data is read, not asked.\n2. **Frame** - state the real decision the artifact must serve and its completion bar.\n3. **Draft** - apply the domain framework. Smallest useful artifact, real inputs, assumptions\n   labeled.\n4. **Critic** - an independent pass red-teams the draft: missing sections, unsourced claims,\n   unvalidated assumptions, dodged decisions - each emitted as a risk.\n5. **Deliver** - fold the risks back in and stop, reporting what was checked.\n\n## Domains (9 pm-skills plugins + screen-design + flow + package = 12)\n\nThe four build-planning domains form a chain on one source of truth (the 기능명세 / SPEC):\n**EXECUTE -> FLOW -> STORYBOARD -> PACKAGE**, contract in [`reference/model.md`](reference/model.md).\n\n| Domain | Covers |\n|---|---|\n| **DISCOVER** | ideation, risky assumptions, opportunity solution tree, feature prioritization, customer interviews, metrics |\n| **STRATEGY** | strategy canvas, vision, value proposition, lean/business model, monetization, pricing, SWOT/PESTLE/Porter/Ansoff |\n| **EXECUTE** | PRD, OKRs, outcome roadmap, sprint, retro, pre-mortem, stories, stakeholder map, prioritization frameworks, red-team |\n| **FLOW** | user-flow graph - nodes derived from SPECs (R2), page->screen hand-off (R3), zero-dep visualization (text list + boxes) |\n| **STORYBOARD** | 화면설계서/기능명세 screen design docs - wireframe (plan-first) or replica/Figma, per-element spec; runs standalone via bundled HTML, optional `storyboard-spec` accelerator |\n| **PACKAGE** | integrated implementation bundle - PRD+spec+flow+wireframe under one snapshotId (R6), back-link integrity, md/json/svg/html export |\n| **RESEARCH** | personas, segmentation, customer journey, market sizing (TAM/SAM/SOM), competitor analysis, sentiment - grounded in live, keyless market signal (voice of customer) |\n| **ANALYTICS** | NL->SQL, cohort analysis, A/B test read-out |\n| **GTM** | go-to-market, beachhead, ICP, growth loops, motions, battlecard |\n| **GROWTH** | marketing ideas, positioning, value-prop statements, naming, North Star metric |\n| **TOOLKIT** | resume review/tailoring, NDA, privacy policy, proofreading |\n| **AI-SHIP** | shipping artifacts, intended-vs-implemented gap (cited evidence) |\n\n## Why a critic\n\nPM artifacts have no unit tests, so a plausible-looking draft can be wrong. `superpm` borrows\n`supergoal`'s discipline: an independent reader checks the artifact against three grounds -\n**framework completeness, the real decision, and the evidence behind every claim** - and names\neach gap as a risk before you ship it. Unsourced numbers and dodged decisions are blockers.\n\n## Install\n\nThis repo **is** the skill. Put it where your agent CLI finds skills:\n\n```bash\ngit clone <repo-url> superpm-skill\nln -s \"$(pwd)/superpm-skill\" <your-agent-skills-dir>/superpm\n# examples: ~/.claude/skills/superpm, ~/.codex/skills/superpm\n```\n\nThen in your agent CLI: `/superpm <your PM ask>`.\n\n## Layout\n\n```\nSKILL.md      thin spine: principles, intent-capture table, the loop, reference map\nreference/    intent · critic · signal · model · discover · strategy · execute · flow · storyboard · package · research · analytics · gtm · growth · toolkit · ai-ship\ntemplates/    PRD · strategy canvas · opportunity solution tree · battlecard · storyboard page/board · workspace (3-tab)\ndocs/         DESIGN.md\n```\n\n## Credit\n\n- PM framework catalog: **[phuryn/pm-skills](https://github.com/phuryn/pm-skills)** (MIT),\n  curated by Paweł Huryn - merged here into one skill.\n- Workflow + verification discipline adapted from **[cskwork/supergoal-skill](https://github.com/cskwork/supergoal-skill)** (MIT).\n- Screen-design storyboard method (optional accelerator) from **[cskwork/storyboard-spec](https://github.com/cskwork/storyboard-spec)** (MIT).\n- Live-signal method (optional `last30days` accelerator) from **[mvanhorn/last30days-skill](https://github.com/mvanhorn/last30days-skill)** (MIT).\n- Keyless read techniques (Jina Reader, RSS) from **[Panniantong/Agent-Reach](https://github.com/Panniantong/Agent-Reach)** (MIT).\n\n## License\n\nMIT. See [`LICENSE`](LICENSE).\n",
  "bytes": 5695,
  "sha": "80a2069b59af1fe6f656f8811e65448a02ddaac5193c6220ed95e73995deb23c",
  "repo_slug": "cskwork/superpm-skill",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_cskwork_superpm_skill_06358559/readme"
}