Discover (5 skills)
Research, competitive analysis, stakeholder mapping.
PM Skills
63 best-practice product management skills for AI agents.
PM Skills teaches AI assistants how to produce professional PM artifacts - PRDs, user stories, acceptance criteria, experiment designs, sprint outputs, and more. One command, consistent output, every time.
The Triple Diamond
Skills are organized across 6 phases of the Triple Diamond framework - three diamonds covering the problem space, the solution space, and the learning space.
graph LR
subgraph "Problem Space"
D1["Discover<br/>5 skills"] --> D2["Define<br/>5 skills"]
end
subgraph "Solution Space"
D3["Develop<br/>4 skills"] --> D4["Deliver<br/>6 skills"]
end
subgraph "Learning Space"
D5["Measure<br/>6 skills"] --> D6["Iterate<br/>4 skills"]
end
D2 --> D3
D4 --> D5
D6 -.->|"next cycle"| D1
classDef discover fill:#e7f5ff,stroke:#1971c2,color:#0c2d5e,stroke-width:2px
classDef define fill:#e6fcf5,stroke:#0ca678,color:#0a4f3c,stroke-width:2px
classDef develop fill:#f3e8ff,stroke:#7048e8,color:#3a1d8a,stroke-width:2px
classDef deliver fill:#fff4e1,stroke:#d9480f,color:#5e2200,stroke-width:2px
classDef measure fill:#fff9db,stroke:#e8a317,color:#5e3e00,stroke-width:2px
classDef iterate fill:#ffe9ec,stroke:#c2255c,color:#5c0a25,stroke-width:2px
class D1 discover
class D2 define
class D3 develop
class D4 deliver
class D5 measure
class D6 iterate
Learn about the Triple Diamond
The Skills
Define (5 skills)
Problem framing, hypotheses, opportunity trees, JTBD.
Develop (4 skills)
Solution briefs, ADRs, design rationale, spikes.
Deliver (6 skills)
PRDs, user stories, acceptance criteria, edge cases, launch, release notes.
Measure (6 skills)
Experiments, instrumentation, dashboards, results, OKR grading.
Iterate (4 skills)
Retrospectives, lessons, refinement, pivot decisions.
Foundation (8 skills)
Cross-cutting skills: persona, OKR writer, lean canvas, meeting lifecycle, stakeholder update.
Utility (10 skills)
Create, validate, iterate skills, generate diagrams and presentations, and update the library.
Tool (15 skills)
Canonical sprint methodologies: Foundation Sprint (7) + Design Sprint (7) + tool-note-and-vote.
Skills by Phase
30 domain skills across 6 phases, plus 8 foundation, 10 utility (including the v2.16.0 sub-agent dispatch slate), and 15 tool skills (Foundation Sprint + Design Sprint families + note-and-vote):
block-beta
columns 6
block:discover["Discover"]:1
d1["/competitive-analysis"]
d2["/interview-synthesis"]
d3["/stakeholder-summary"]
d4["/journey-map"]
d5["/market-sizing"]
end
block:define["Define"]:1
f1["/problem-statement"]
f2["/hypothesis"]
f3["/opportunity-tree"]
f4["/jtbd-canvas"]
f5["/prioritization-framework"]
end
block:develop["Develop"]:1
v1["/solution-brief"]
v2["/spike-summary"]
v3["/adr"]
v4["/design-rationale"]
end
block:deliver["Deliver"]:1
l1["/prd"]
l2["/user-stories"]
l3["/acceptance-criteria"]
l4["/edge-cases"]
l5["/launch-checklist"]
l6["/release-notes"]
end
block:measure["Measure"]:1
m1["/experiment-design"]
m2["/instrumentation-spec"]
m3["/dashboard-requirements"]
m4["/experiment-results"]
m5["/okr-grader"]
m6["/survey-analysis"]
end
block:iterate["Iterate"]:1
i1["/retrospective"]
i2["/lessons-log"]
i3["/refinement-notes"]
i4["/pivot-decision"]
end
discover --> define --> develop --> deliver --> measure --> iterate
style discover fill:#e7f5ff,stroke:#1971c2,color:#0c2d5e,stroke-width:2px
style define fill:#e6fcf5,stroke:#0ca678,color:#0a4f3c,stroke-width:2px
style develop fill:#f3e8ff,stroke:#7048e8,color:#3a1d8a,stroke-width:2px
style deliver fill:#fff4e1,stroke:#d9480f,color:#5e2200,stroke-width:2px
style measure fill:#fff9db,stroke:#e8a317,color:#5e3e00,stroke-width:2px
style iterate fill:#ffe9ec,stroke:#c2255c,color:#5c0a25,stroke-width:2px
Plus: /lean-canvas /persona /okr-writer /meeting-agenda /meeting-brief /meeting-recap /meeting-synthesize /stakeholder-update (Foundation - cross-cutting) · /pm-skill-builder /pm-skill-validate /pm-skill-iterate /mermaid-diagrams /slideshow-creator /update-pm-skills /pm-critic /pm-audit-repo /pm-draft-changelog /pm-release (Utility incl. v2.16.0 sub-agent dispatch slate) · /tool-foundation-sprint-readiness /tool-foundation-sprint-brief /tool-foundation-sprint-basics /tool-foundation-sprint-differentiation /tool-foundation-sprint-approach-options /tool-foundation-sprint-magic-lenses /tool-foundation-sprint-founding-hypothesis (Foundation Sprint family) · /tool-design-sprint-readiness /tool-design-sprint-brief /tool-design-sprint-map-and-target /tool-design-sprint-sketch /tool-design-sprint-decide-and-storyboard /tool-design-sprint-prototype-plan /tool-design-sprint-test-and-score (Design Sprint family) · /tool-note-and-vote (standalone)
The Skill Lifecycle
Three utility skills form a self-reinforcing quality loop for managing the skill library itself:
flowchart LR
Create["/pm-skill-builder<br/>Create"] --> Validate["/pm-skill-validate<br/>Validate"]
Validate --> Decision{Findings?}
Decision -- "PASS" --> Ship["Ship"]
Decision -- "WARN / FAIL" --> Iterate["/pm-skill-iterate<br/>Iterate"]
Iterate --> Validate
Create a new skill with guided gap analysis and classification. Validate it against structural conventions and quality criteria. Iterate to fix findings from the validation report or apply feedback. Repeat until passing, then ship.
The lifecycle tools are what keep the library consistent as it grows - the validator catches drift, and the iterator applies fixes with version tracking and change summaries.
Learn more about the lifecycle · PM-Skill versioning
Quick Start
git clone https://github.com/product-on-purpose/pm-skills.gitcd pm-skillsThen use any skill:
/prd "Search feature for e-commerce platform"/hypothesis "Will one-page checkout increase conversion?"/acceptance-criteria "User can reset password via email"Full setup guide · Find the right skill · Recipes
See It In Action
Follow three fictional companies through the complete product lifecycle - from discovery research to pivot decisions - with real prompts and full outputs.
Storevine - B2B Ecommerce
Building email marketing for 15K merchants. Organized prompts.
Brainshelf - Consumer PKM
Building a morning digest for 22K users. Casual prompts.
Workbench - Enterprise Collaboration
Building document templates for 500 enterprises. Structured prompts.
Works Everywhere
| Platform | Method |
|---|---|
| Claude Code | Slash commands (/prd, /hypothesis, etc.) |
| GitHub Copilot | AGENTS.md auto-discovery |
| Cursor / Windsurf | AGENTS.md or MCP server |
| Claude.ai / Desktop | ZIP upload or MCP |
| Any MCP client | pm-skills-mcp |
Workflows
12 guided multi-skill workflows for common PM processes. Each chains skills in a recommended sequence with handoff guidance.
| Workflow | Best for | Skills |
|---|---|---|
| Feature Kickoff | New features | problem-statement, hypothesis, prd, user-stories, launch-checklist |
| Lean Startup | Rapid validation | hypothesis, experiment-design, experiment-results, pivot-decision |
| Triple Diamond | Major initiatives | All 30 phase skills across 6 phases |
| Customer Discovery | Research to problem | interview-synthesis, jtbd-canvas, opportunity-tree, problem-statement |
| Sprint Planning (agile) | Agile sprint-ready stories | refinement-notes, user-stories, edge-cases |
| Product Strategy | Strategic framing | competitive-analysis, stakeholder-summary, opportunity-tree, solution-brief, adr |
| Post-Launch Learning | Ship to learn | instrumentation-spec, dashboard-requirements, experiment-results, retrospective, lessons-log |
| Stakeholder Alignment | Leadership buy-in | stakeholder-summary, problem-statement, solution-brief, launch-checklist |
| Technical Discovery | Feasibility | spike-summary, adr, design-rationale |
| Foundation Sprint | 2-day strategic alignment | 7 tool-foundation-sprint-* skills producing a testable Founding Hypothesis |
| Design Sprint | 5-day prototype + test | 7 tool-design-sprint-* skills producing a Decider’s build/iterate/pivot/stop call |
| Foundation to Design | End-to-end FS + DS arc | Both families chained with a narrative handoff conversation |
Recent Releases
| Version | Date | Highlights |
|---|---|---|
| v2.18.0 | 2026-05-21 | Highest-consensus PM skill gaps: 4 new phase skills (discover-market-sizing, define-prioritization-framework, discover-journey-map, measure-survey-analysis), each with a companion command and 3 thread samples. Catalog 59 to 63 (phase 26 to 30); commands 66 to 70. |
| v2.17.0 | 2026-05-20 | Native Claude Code sub-agent registration (definitions moved to the canonical agents/ directory; coordination dir renamed AGENTS/ to _agent-context/); frontmatter metadata-nested migration; bash-3.2 validator portability. 59-skill catalog unchanged. |
| v2.16.0 | 2026-05-17 | Active Orchestration + Doc-Stack Modernization: 4 plugin sub-agents (pm-critic, pm-skill-auditor, pm-changelog-curator, pm-release-conductor) codify the Phase 0 Adversarial Review Loop + 6-gate release runbook including the new G2.5 commit gate. 4 dispatch skills at skills/utility-pm-{role}/ deliver cross-client compatibility (VALIDATED on Codex CLI 2026-05-17 per GATE B + C). Astro 5.13.x to 6.3.x upgrade + Starlight 0.39.x + Node 22.12+ across 5 CI workflows; closes 2 deferred Dependabot alerts. 10 new public docs (concepts, guides, reference, contributing - including the canonical sub-agent compatibility matrix and CI overview) + 12 sub-agent library samples across 3 narrative threads. Content catalog 55 to 59 with the 4 utility dispatch skills; new shipping unit is the sub-agent |
| v2.15.2 | 2026-05-17 | v2.15.x Cycle Closeout + v2.16.0 Plan Reconciliation: closeout patch successor to v2.15.1. No source-code, validator behavior, or catalog changes. Planning-doc hygiene only: audit doc status flipped from DRAFT to REMEDIATION SHIPPED with finding-by-finding closure table; v2.15.0 master plan continuity updated; v2.16.0 plan slate reconciled against v2.15.1 shipped reality (CONTEXT.md prereq marked DONE; ci-plan validator scope reduced via carry-in reconciliation section); CONTEXT.md refreshed; version surfaces bumped |
| v2.15.1 | 2026-05-16 | Post-Tag Audit Remediation + Preventive CI: same-day patch closing 18 v2.15.x audit findings (docs-site homepage, skills landing page, AGENTS.md command table, sample-library README, Sprint Planning naming-discipline) plus a workflow-generator bug fix and 4 new preventive CI validators (landing-page count assertion, workflow-generator coverage, AGENTS.md command-sync, pre-tag validator bundle orchestration). 55-skill catalog unchanged |
| v2.15.0 | 2026-05-16 | Sprint Skills Launch: 15 new skills under the new classification: tool taxonomy (7 Foundation Sprint family + 7 Design Sprint family + 1 tool-note-and-vote standalone) implementing Knapp/Zeratsky/Kowitz canonical sprint methodologies. Skill catalog 40 to 55. Two family contracts with enforcing CI validators (each with —strict release mode). 3 new workflows including end-to-end FS-to-DS arc with narrative handoff conversation that replaces dropped bridge skill (12-row slot-mapping table + 3-question go/no-go checkpoint). 45 library samples across 3 narrative threads (Brainshelf, Storevine, Workbench). Two user guides + two concept docs. Two adversarial-review cycles documented (FS-track + DS-track; all P1 closed pre-tag) |
| v2.14.2 | 2026-05-10 | Codex final review closure (cumulative docs hygiene patch). 0 P0, 1 P1, 11 P2, 1 P3 findings addressed. validate-docs-frontmatter scope expanded to .mdx (parity with V6); check-no-body-h1 doc clarified with “what is NOT caught” framing; validate-mcp-sync guide refreshed for observe-mode default; sync-agents-md.yml workflow_dispatch hardened with two-layer defense (input gate + token gate); pm-skills-mcp README cross-repo update (5 “25 skills” residues to “40 skills”; v2.9.3 latest pointer); CONTRIBUTING.md workaround count corrected; release plan + Release_v2.14.0 deferral table reframed. 40-skill catalog unchanged |
| v2.14.1 | 2026-05-10 | Polish + V15 regression fix: same-day patch for v2.14.0. Title duplication fixed across 62 hand-authored docs + 6 generator emission sites (Starlight title-vs-body-H1 migration regression); 63 generator pages cleaned of contributor-noise asides; 182 en-dashes swept across 45 library samples; Mermaid 3-layer beautification + canonical style guide; two validators promoted to truly enforcing + new check-no-body-h1 validator added (enforcing tier 11 to 14); MCP maintenance posture codified cross-repo |
| v2.14.0 | 2026-05-10 | Doc Stack Migration: MkDocs Material to Astro Starlight: doc-stack migration release with zero new skills (40 unchanged); 4 phases / 13 workstreams; production cutover via GitHub Pages source flip to Actions; Starlight-native asides + MDX details replace pymdownx admonitions; client-side Mermaid via astro-mermaid 2.0.1; 115 library samples mounted under /samples/; 12 redirect entries preserved with /pm-skills/ base path; Phase 0 Adversarial Review Loop via codex:rescue against trunk |
| v2.13.0 | 2026-05-05 | Foundation Hardening + Doc Stack Decision: maintenance and quality release with zero new skills; 7 new CI gates that catch doc drift on PRs (validator inventory 15 to 22; enforcing tier 5 to 10); Diataxis-aligned docs with pm-skill-* filename prefix; Pattern 5C generated-content marker; out-of-cycle pm-skills-mcp v2.9.3 security patch |
| v2.12.0 | 2026-05-03 | OKR Skills Launch: foundation-okr-writer + measure-okr-grader with canonical type and indicator-class enums; 38 to 40 skills |
| v2.11.1 | 2026-04-22 | skills.sh CLI compatibility patch: unblocks npx skills add product-on-purpose/pm-skills; lint hardening; em-dash sweep completion |
| v2.11.0 | 2026-04-18 | Meeting Skills Family: 5 foundation skills + canonical contract + enforcing CI; lean-canvas; 32 to 38 skills |
| v2.10.0 | 2026-04-11 | Utility skill expansion: /mermaid-diagrams, /slideshow-creator, /update-pm-skills. (v2.10.1 and v2.10.2 patches followed; see CHANGELOG.md for details.) |
| v2.9.1 | 2026-04-10 | Workflows guide + docs count consistency CI |
| v2.9.0 | 2026-04-06 | Workflows: rename bundles to workflows, expand 3 to 9, 7 /workflow-* commands |