{
  "markdown": "<p align=\"center\">\n  <picture>\n    <source media=\"(prefers-color-scheme: dark)\" srcset=\"assets/banner.png\">\n    <img src=\"assets/banner.png\" width=\"100%\" alt=\"Product Superpowers banner\">\n  </picture>\n</p>\n\n<p align=\"center\">\n  <strong>Give your AI agent a PM brain.</strong>\n</p>\n\n<p align=\"center\">\n  <a href=\"#quick-start\">Quick Start</a> ·\n  <a href=\"#how-it-works\">How It Works</a> ·\n  <a href=\"#skills\">Skills</a> ·\n  <a href=\"#faq\">FAQ</a>\n</p>\n\n---\n\n## What is this?\n\nProduct Superpowers is a set of 18 composable skills that give your AI agent a complete product management methodology — including subagent-powered research, review, and visual collaboration. Instead of jumping straight to writing requirements, your agent runs discovery, validates problems, writes PRDs, breaks them into stories, prioritizes, and plans launches — systematically, every time.\n\n### The pipeline\n\n```\nIdea  →  Discovery  →  PRD  →  Stories  →  Prioritize  →  Launch\n          ↕             ↕        ↕            ↕               ↕\n      Roadmap   ·   Analytics  ·  Stakeholders  ·  Competitive  ·  Strategy  ·  Continuous Discovery\n```\n\n## Full PDLC coverage\n\nCovers the complete Product Development Life Cycle — from ideation to commercialization:\n\n| PDLC Stage | Skill |\n|-----------|-------|\n| 1. Ideation | `product-discovery` — JTBD, user interviews, problem validation |\n| 2. Idea screening | `prioritization` — RICE, ICE, MoSCoW scoring |\n| 3. Concept testing | `continuous-discovery` — assumption testing, concierge MVP |\n| 4. Business analysis | `product-strategy` + `competitive-analysis` — PMF, SWOT, build vs buy |\n| 5. Product design | `design-handoff` + `user-story-writing` — tokens, states, INVEST, Gherkin |\n| 6. Market testing | `launch-planning` — beta programs, staged rollout |\n| 7. Commercialization | `launch-planning` — GTM strategy, launch checklists |\n\n## Why this?\n\n| If you... | Your agent will... |\n|-----------|-------------------|\n| Have an idea for a feature | Run discovery first — JTBD, user interviews, opportunity assessment. Never jumps to solution. |\n| Need a PRD written | Outputs Amazon PR/FAQ format — press release + external FAQ + internal FAQ. Metrics, assumptions, constraints, all included. |\n| Have a PRD ready | Breaks it into INVEST-validated user stories with Gherkin acceptance criteria. Every state documented. |\n| Need to prioritize | Scores with RICE, segments with MoSCoW, classifies with Kano. Outputs ordered list with rationale. |\n| Are about to launch | Produces pre/during/post-launch checklists, beta plan, rollout strategy, kill criteria, and post-launch rituals. |\n| Want to keep discovery alive | Sets up weekly interview cadence, Opportunity Solution Trees, and assumption testing — Teresa Torres framework. |\n\n---\n\n## Quick Start\n\n### Claude Cowork (recommended for PMs)\n\n[Claude Cowork](https://claude.com/product/cowork) is Anthropic's AI agent for knowledge work — no terminal needed. It runs in the Claude desktop app, reads your files, opens your browser, and completes PM tasks autonomously.\n\n**Install via Git (local):**\n\n```bash\n# Clone to your local machine\ngit clone https://github.com/guhcostan/product-superpowers.git ~/product-superpowers\n```\n\nThen in the Claude Desktop app:\n1. Switch to the **Cowork** tab\n2. Click **Customize** in the left sidebar\n3. Click **Upload plugin** and select the `~/product-superpowers` folder\n4. Verify: ask Claude **\"What skills do you have for product management?\"**\n\n> **Why Cowork for PMs?** Unlike Claude Code (terminal, coding-focused), Cowork works in the desktop app — open your PRDs in Notion, research competitors in Chrome, pull data from spreadsheets, and Claude orchestrates it all. It's built for the kind of work PMs actually do.\n\n### Claude Code (terminal)\n\n```bash\n/plugin marketplace add obra/superpowers-marketplace\n/plugin install product-superpowers@superpowers-marketplace\n```\n\n### OpenCode\n\nAdd to `opencode.json`:\n\n```json\n{ \"plugin\": [\"product-superpowers@git+https://github.com/guhcostan/product-superpowers.git\"] }\n```\n\nVerify: **\"Tell me about your product superpowers\"**\n\n---\n\n## Skills\n\n<details>\n<summary><strong>Core Workflow</strong> — the primary pipeline</summary>\n\n| Skill | What triggers it |\n|-------|-----------------|\n| `product-discovery` | New idea, feature request, \"build X\" |\n| `writing-prd` | Discovery approved, need requirements |\n| `user-story-writing` | PRD approved, need development stories |\n\n</details>\n\n<details>\n<summary><strong>Decision Making</strong> — what to build and when</summary>\n\n| Skill | What triggers it |\n|-------|-----------------|\n| `prioritization` | \"Prioritize\", backlog grooming |\n| `roadmapping` | \"Roadmap\", OKRs, quarterly planning |\n\n</details>\n\n<details>\n<summary><strong>Execution</strong> — shipping with quality</summary>\n\n| Skill | What triggers it |\n|-------|-----------------|\n| `design-handoff` | \"Handoff\", design specs for engineering |\n| `launch-planning` | \"Launch\", \"release\", \"GTM\" |\n| `stakeholder-management` | Status updates, exec summaries |\n| `product-analytics` | Metrics, KPIs, A/B tests |\n| `pm-feedback-synthesis` | Interviews, surveys, tickets → themes & insights |\n\n</details>\n\n<details>\n<summary><strong>Continuous Strategy</strong> — the long game</summary>\n\n| Skill | What triggers it |\n|-------|-----------------|\n| `competitive-analysis` | \"Competitors\", SWOT, market analysis |\n| `product-strategy` | Vision, build vs buy, PMF |\n| `continuous-discovery` | Discovery cadence, OSTs, assumption testing |\n\n</details>\n\n<details>\n<summary><strong>Agentic</strong> — subagent-powered workflows</summary>\n\n| Skill | What triggers it |\n|-------|-----------------|\n| `pm-parallel-research` | Multiple independent research tasks |\n| `pm-artifact-review` | PRD/stories/roadmap ready for review |\n| `pm-visual-workspace` | Show OSTs, story maps, roadmaps visually |\n| `pm-autonomous-execution` | PM plan ready, automatic execution |\n\n</details>\n\n<details>\n<summary><strong>Meta</strong></summary>\n\n| Skill | What triggers it |\n|-------|-----------------|\n| `using-product-superpowers` | Every session — bootstrap |\n\n</details>\n\n---\n\n## How It Works\n\n1. **You say \"build X\"** → Agent loads `product-discovery`. Asks you JTBD questions. Maps opportunities. Validates before any requirements.\n\n2. **You approve discovery** → Agent loads `writing-prd`. Writes a press release, customer FAQ, internal FAQ. Defines metrics and scope.\n\n3. **You approve the PRD** → Agent loads `user-story-writing`. Breaks it into INVEST stories with Given-When-Then acceptance criteria.\n\n4. **Parallel support skills** trigger as needed: `prioritization` when you ask what to build first, `roadmapping` when you update the plan, `launch-planning` when you're close to shipping, `product-analytics` when you define metrics.\n\n---\n\n## Philosophy\n\n- **Discovery before requirements** — Understand the problem before writing what to build\n- **Outcomes over outputs** — Measure success by business impact, not features shipped\n- **Evidence over opinions** — Validate with data and user research\n- **Systematic over ad-hoc** — Process over guessing\n\n---\n\n## FAQ\n\n**Does this replace a human PM?**\nNo. Product Superpowers automates the systematic parts of PM work — discovery structure, PRD formatting, story breakdown. Strategic judgment, taste, and deep user empathy remain human.\n\n**Do I have to follow the full pipeline?**\nThe three core skills (discovery → PRD → stories) have hard gates — they won't let you skip steps. All other skills are flexible and invoked on demand.\n\n**Does this work with Jira, Linear, Notion?**\nThe skills produce structured outputs that maps naturally to any tool. Specific integrations come later.\n\n**What PM frameworks does it use?**\nAmazon PR/FAQ, SVPG Opportunity Assessment, Teresa Torres' OSTs, Shreyas Doshi's frameworks, RICE/ICE/MoSCoW/Kano prioritization, JTBD, Gherkin acceptance criteria, and more.\n\n**Can I use this with Claude Cowork, Claude Code, or OpenCode?**\nProduct Superpowers is built for Claude Cowork (recommended for PMs — desktop app, no terminal). Also works in Claude Code (terminal) and OpenCode. See Quick Start for each.\n\n---\n\n## References\n\nProduct Superpowers draws from: Marty Cagan (SVPG), Teresa Torres, Shreyas Doshi, Amazon PR/FAQ, Clayton Christensen (JTBD), John Doerr (OKRs), Melissa Perri, Rob Fitzpatrick (The Mom Test), April Dunford, and Strategyzer.\n\n---\n\nMIT · [Contribute](https://github.com/guhcostan/product-superpowers) · [Issues](https://github.com/guhcostan/product-superpowers/issues)\n",
  "bytes": 8538,
  "sha": "30ae6c7748abda8a0682047ec07b30974657e3626a671436d79cc59b2fe646f7",
  "repo_slug": "guhcostan/product-superpowers",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_guhcostan_product_superpowers_a6f90d9a/readme"
}