The PM operating system for AI coding assistants. Discovery → delivery → career, with Jira, Linear, Notion, Confluence integrations.
★ The deepest and most-used domain in Claude Skills. PMs from APM to CPO use these skills daily to ship faster with fewer meetings.
- 54 production-ready skills covering every stage: discovery, definition, execution, delivery, launch, career growth
- Portable artifacts — generate once, push to Jira, Linear, Notion, Confluence, or GitHub Projects
- Real automation, not prompts — 15+ Python tools you can pipe into CI, scripts, MCP servers
- Exec-ready outputs — PRDs, roadmaps, status updates that don't read like AI slop
- Career track included — interview prep, career ladder, onboarding, 1:1s (no PM tool covers this)
New to the library? Pick your role:
execution/create-prd/— Write your first PRD with structureexecution/wwas/— Why-What-Acceptance backlog itemscareer/pm-onboarding/— 30-60-90 day plancareer/pm-interview-prep/— Land your next role
discovery/brainstorm-ideas/→discovery/identify-assumptions/— Product Trio discoveryexecution/prioritization-frameworks/— RICE, ICE, MoSCoW, Opportunity Scoreexecution/status-update-generator/— Weekly exec update from Jira/Linear dataexamples/feature-end-to-end.md— Idea → release notes in 6 commands
senior-pm/— Portfolio management, stakeholder mapping, EMV risk analysisexecution/north-star-metric/— NSM + input metric treeexecution/roadmap-communication/— Exec / customer / internal roadmap variantsexecution/daci-framework/— Decision facilitation at scale
scrum-master/— Sprint analytics, velocity forecasting, capacity planningexecution/backlog-refinement/— INVEST + DoR/DoD + splitting heuristicsexecution/story-splitting/— 9 splitting patterns (Lawrence)sprint-retrospective/— Data-driven retros
delivery-manager/— Release management, deployment coordinationexecution/launch-playbook/— Internal + external launch coordinationexecution/cycle-time-analyzer/— Flow metrics (Little's Law, CFD)execution/release-notes/— User-facing release comms
program-manager/— Multi-project coordination, dependency managementexecution/dependency-map/— Cross-team blockers, critical pathexecution/daci-framework/— Cross-team decision governanceexecution/roadmap-communication/— Stakeholder-specific roadmaps
execution/north-star-metric/— Define the NSM + input treeexecution/outcome-roadmap/— Output → outcome transformationexecution/brainstorm-okrs/— OKR brainstorming (Wodtke)career/pm-career-ladder/— Rubrics for your team
Examples of what these skills produce. (Click to enlarge — screenshots coming, see examples/ for raw outputs.)
| Artifact | Skill | Preview |
|---|---|---|
| 8-section PRD | execution/create-prd/ |
Markdown, exec-ready |
| RICE-scored backlog | execution/prioritization-frameworks/ |
Sortable table + rationale |
| User story map | execution/story-mapping/ |
Mermaid journey diagram |
| Cumulative Flow Diagram | execution/cycle-time-analyzer/ |
Mermaid gantt |
| Opportunity Solution Tree | discovery/interview-synthesis/ |
Mermaid graph LR |
| Cross-team dependency map | execution/dependency-map/ |
Mermaid graph LR |
| Weekly exec status update | execution/status-update-generator/ |
Markdown, traffic-light |
| Pre-mortem risk matrix | discovery/pre-mortem/ |
Tiger/Paper Tiger/Elephant |
| Skill | Focus | Python Tools |
|---|---|---|
senior-pm/ |
Portfolio, stakeholder mapping, EMV risk | 4 scripts |
scrum-master/ |
Sprint analytics, velocity, team health | 4 scripts |
delivery-manager/ |
Release, deployment, incident response | — |
program-manager/ |
Multi-project coordination, dependencies | — |
agile-coach/ |
Agile transformation, maturity assessment | — |
jira-expert/ |
Jira admin, JQL, automation | — |
confluence-expert/ |
Documentation, knowledge management | — |
atlassian-admin/ |
Suite administration | — |
atlassian-templates/ |
Ready-to-use Jira/Confluence templates | — |
sprint-retrospective/ |
Data-driven retros | — |
| Skill | Focus |
|---|---|
linear-expert/ |
Linear admin, GraphQL queries, Jira → Linear migration |
notion-pm/ |
Notion DBs for PRDs/OKRs/Roadmap/Decisions, API patterns |
productboard-expert/ ★ NEW |
Productboard admin, Insight inbox triage, Driver scoring, REST API |
| Skill | Focus | Framework |
|---|---|---|
brainstorm-ideas/ |
Product Trio ideation | Opportunity Solution Tree (Torres) |
brainstorm-experiments/ |
Lean experiment design | XYZ Hypothesis (Savoia) |
identify-assumptions/ |
Assumption mapping | Torres 4-8 risk categories |
pre-mortem/ |
Pre-launch risk analysis | Tiger/Paper Tiger/Elephant (Klein) |
interview-synthesis/ |
Interview → opportunity tree | Torres + Portigal + Klement |
customer-interview-script/ ★ NEW |
How to RUN a discovery interview | Portigal + Fitzpatrick + Torres |
value-proposition-canvas/ ★ NEW |
Strategyzer VPC (Customer Profile + Value Map) | Osterwalder + Strategyzer |
jtbd-workshop/ ★ NEW |
Full Jobs-To-Be-Done workshop | Christensen + Ulwick + Moesta + Klement |
| Skill | Focus | Framework |
|---|---|---|
create-prd/ |
8-section PRD scaffolding | PRD canonical |
prfaq/ |
Working Backwards PR/FAQ | Amazon Bezos PRFAQ |
ai-feature-prd/ ★ NEW |
PRDs for AI/ML features | Karpathy + Anthropic RSP + Reforge AI PM |
pricing-prd/ ★ NEW |
Pricing experiments and pricing-page PRDs | Ramanujam + Campbell + Westendorp PSM |
product-vision/ ★ NEW |
Vision document (above NSM) | Pichler + Moore + Raskin + Cagan |
brainstorm-okrs/ |
OKR brainstorming | Wodtke "Radical Focus" |
north-star-metric/ |
NSM + input metric tree | Ellis + Amplitude framework |
quarterly-planning/ ★ NEW |
Full Q1 cycle (kickoff → mid-Q → close) | Wodtke + Cagan + Reforge + RAD |
outcome-roadmap/ |
Output → outcome roadmap | Now/Next/Later |
roadmap-communication/ |
Exec/customer/internal variants | Cagan right-sizing |
prioritization-frameworks/ |
9-framework scoring | RICE, ICE, Opportunity, MoSCoW, Weighted |
backlog-refinement/ |
INVEST + DoR/DoD + splitting | INVEST (Wake) |
story-splitting/ |
Vertical slicing patterns | Lawrence 9 patterns |
story-mapping/ |
User story mapping | Jeff Patton |
job-stories/ |
JTBD backlog format | When/Want/So (Klement) |
wwas/ |
Why-What-Acceptance backlog | WWAS |
customer-feedback-triage/ ★ NEW |
Inbound feedback → categorize → score → backlog | Kano + Cagan request-to-opportunity |
activation-funnel/ ★ NEW |
AARRR funnel design + analyzer | McClure + Ellis + Chen + Reforge |
feature-flag-strategy/ ★ NEW |
Phased rollouts, kill-switches, flag debt | Fowler toggle taxonomy + Reforge |
cycle-time-analyzer/ |
Flow metrics, CFD, Little's Law | Vacanti Kanban metrics |
dependency-map/ |
Cross-team blockers, critical path | CPM + DSM + Conway's Law |
status-update-generator/ |
Weekly exec update | SBNR + traffic-light |
summarize-meeting/ |
Meeting summaries | Action items + decisions |
daci-framework/ |
Decision facilitation | DACI |
beta-program/ |
Closed beta playbook | Kano + cohort design |
launch-playbook/ |
Launch coordination | Alpha/beta/GA + RACI |
post-mortem/ ★ NEW |
Blameless incident RCA | Google SRE + Allspaw + Dekker + Perrow |
release-notes/ |
Release notes from tickets | Category-based |
eol-communication/ |
End-of-life messaging | Sunset playbook |
| Skill | Focus | Framework |
|---|---|---|
pm-interview-prep/ |
APM → Group PM interview prep | CIRCLES (Lin) + STAR + Decode and Conquer |
pm-career-ladder/ |
Rubrics + growth + promo packets | Square/Stripe ladders + Reforge |
pm-onboarding/ |
30-60-90 day plan | Watkins "First 90 Days" + STARS |
pm-1on1s/ |
1:1 templates by partner type | Radical Candor + GROW + Manager Tools |
★ NEW = added in the 2026-05-22 expansion (12 new skills: 3 discovery, 8 execution, 1 integration)
Every skill ships with a worked example. Inside each skill folder is examples/<scenario>.md with a realistic scenario, the skill's workflow applied, and the finished artifact (full PRD, full post-mortem, full status update, etc.) PMs can copy and adapt.
Plus end-to-end multi-skill scenarios in the top-level examples/:
feature-end-to-end.md— Idea → release notes in 6 commands- More scenarios in
examples/README.md
A few highlight per-skill examples:
| Skill | Example artifact |
|---|---|
execution/create-prd/ |
Full 8-section PRD for "Shared Dashboards" |
execution/ai-feature-prd/ |
PRD for "AI Meeting Notes Summarizer" (eval spec + guardrails + model fallback) |
execution/post-mortem/ |
Blameless RCA for a 47-minute payment outage |
execution/status-update-generator/ |
Weekly exec update for "Acme Search Platform" |
execution/activation-funnel/ |
AARRR diagnosis for a freemium SaaS with poor day-7 retention |
discovery/interview-synthesis/ |
8-interview synthesis → opportunity solution tree |
career/pm-interview-prep/ |
Full CIRCLES answer for a Stripe product-sense question |
Twelve of the artifact-generating PM skills now have deterministic eval rubrics at evals/ — scores any markdown artifact 0-100 against the skill's red flags and success criteria. No LLM in the loop, stdlib only.
# Score all worked examples
python evals/run.py --all --threshold 70
# Score one skill against its own example
python evals/run.py --skill post-mortem
# Score a custom artifact
python evals/run.py --skill create-prd --artifact ./my-prd.mdSkills with rubrics: create-prd, prfaq, ai-feature-prd, brainstorm-okrs, status-update-generator, post-mortem, north-star-metric, product-vision, pricing-prd, roadmap-communication, release-notes, customer-feedback-triage.
Use case: PMs paste their drafted artifact and get back specific failures ("Summary is 12 sentences, need 2-5", "Section 5 Market Segments not found", "Found forbidden buzzword: synergy"). Faster than a human review, catches form failures consistently.
Every PM skill now ships with a references/red-flags.md file: 10-12 concrete anti-patterns with paired bad and good artifact snippets, plus a "How to catch it" check question. Use it before sharing your output to scan for the most common failures.
Examples of red flags caught:
- status-update-generator: "watermelon" status (green outside, red inside) — when the rationale sentence contradicts the color
- create-prd: solution-before-problem (Section 7 expands while Section 3 is one sentence)
- post-mortem: blame language ("Dev shipped a bad config") vs systemic framing ("Config validation lacked a pre-deploy step")
- brainstorm-okrs: output-as-KR ("Ship the feature") vs outcome-as-KR ("Increase weekly active teams by 30%")
- customer-feedback-triage: squeaky-wheel bias (the loudest customer's request leapfrogs the scored backlog)
See any skill's references/red-flags.md for the full library.
Five pipelines/ at the repo root chain multiple PM skills into single-command flows:
| Pipeline | Chains | Use case |
|---|---|---|
pipelines/feature-end-to-end.py |
identify-assumptions → brainstorm-experiments → pre-mortem → create-prd → brainstorm-okrs → prioritization-frameworks → release-notes | New feature from idea to launch |
pipelines/weekly-cadence.py |
jira/linear adapter → status-update + flow-metrics + dependency-map | Weekly exec status from real tracker data |
pipelines/customer-discovery.py |
interview-synthesis → identify-assumptions → brainstorm-experiments → north-star-metric | Process a batch of customer interviews |
pipelines/post-mortem-flow.py |
post-mortem → pre-mortem (for follow-ups) → dependency-map | After an incident |
pipelines/launch-coordination.py |
beta-program → feature-flag-strategy → launch-playbook → release-notes | Coordinated GA launch |
python pipelines/feature-end-to-end.py --demo --output ./my-feature
# Produces: 01-assumptions.md, 02-experiments.md, 03-pre-mortem.md,
# 04-prd.md, 05-okrs.md, 06-priorities.md, 07-release-notes.md,
# summary.mdPM tools all consume JSON. To remove the "first get the data" friction, ship adapters that pull from real APIs:
# Jira -> status-update-generator
python tools/adapters/jira_to_json.py --jql "sprint in openSprints()" --format status-update \
| python project-management/execution/status-update-generator/scripts/status_generator.py --input -
# Linear -> cycle-time-analyzer
python tools/adapters/linear_to_json.py --team ENG --cycle previous --format cycle-time \
| python project-management/execution/cycle-time-analyzer/scripts/flow_metrics.py --input -
# Notion -> customer-feedback-triage
python tools/adapters/notion_to_json.py --database-id <uuid> --format feedback \
| python project-management/execution/customer-feedback-triage/scripts/feedback_triage.py --input -Stdlib only, env-var auth (JIRA_TOKEN, LINEAR_API_KEY, NOTION_TOKEN). See tools/adapters/README.md.
15 PM skills are exposed as Claude Code MCP tools — callable inside any AI conversation without manually running Python:
pm_create_prd · pm_status_update · pm_funnel_analyze · pm_flow_metrics · pm_dependency_map · pm_feedback_triage · pm_nsm_tree · pm_refinement_score · pm_interview_synthesize · pm_prioritize · pm_okr_validate · pm_roadmap_transform · pm_pre_mortem · pm_release_notes · pm_stakeholder_map
Wire up via .mcp.json:
{ "mcpServers": { "claude-skills": { "command": "python3", "args": ["scripts/mcp_server.py"] } } }| Tool | Skill | Push direction |
|---|---|---|
| Jira | jira-expert/ |
Issues, sprints, JQL, automation |
| Confluence | confluence-expert/ |
Pages, spaces, templates |
| Linear | linear-expert/ |
Issues, cycles, projects, initiatives |
| Notion | notion-pm/ |
DB-driven PRDs, OKRs, roadmaps |
| Productboard | productboard-expert/ |
Insights, Features, Drivers, Releases |
| GitHub Projects | (any execution skill) | Issues via gh CLI |
| Atlassian MCP | (any) | Direct MCP server integration |
Every PM Python tool supports --format json|markdown|mermaid|confluence|notion|linear. See SHARED_OUTPUT_SCHEMA.md.
All tools are CLI-first, standard library only, with --help, --demo, and --format support.
# Discovery
python discovery/brainstorm-experiments/scripts/experiment_designer.py --demo
python discovery/identify-assumptions/scripts/assumption_tracker.py --demo
python discovery/pre-mortem/scripts/risk_categorizer.py --demo
python discovery/interview-synthesis/scripts/interview_synthesizer.py --demo
# Execution
python execution/create-prd/scripts/prd_scaffolder.py --demo
python execution/brainstorm-okrs/scripts/okr_validator.py --demo
python execution/outcome-roadmap/scripts/roadmap_transformer.py --demo
python execution/prioritization-frameworks/scripts/prioritization_scorer.py --demo --framework rice
python execution/release-notes/scripts/release_notes_generator.py --demo
python execution/backlog-refinement/scripts/refinement_scorer.py --demo
python execution/status-update-generator/scripts/status_generator.py --demo
python execution/north-star-metric/scripts/metric_tree_builder.py --demo
python execution/cycle-time-analyzer/scripts/flow_metrics.py --demo
python execution/dependency-map/scripts/dependency_graph.py --demo
# Senior PM
python senior-pm/scripts/stakeholder_mapper.py --demo
python senior-pm/scripts/risk_matrix_analyzer.py --demo
python senior-pm/scripts/project_health_dashboard.py --demo
python senior-pm/scripts/resource_capacity_planner.py --demo
# Scrum Master
python scrum-master/scripts/sprint_capacity_calculator.py --demo
python scrum-master/scripts/velocity_analyzer.py --demo
python scrum-master/scripts/sprint_health_scorer.py --demo
python scrum-master/scripts/retrospective_analyzer.py --demo# Install all PM skills (54)
npx @borghei/claude-skills add pm-starter-pack
# Install individual skills
npx @borghei/claude-skills add senior-pm
npx @borghei/claude-skills add status-update-generator
npx @borghei/claude-skills add pm-interview-prepSupported assistants: Claude Code, Cursor, Codex, Gemini CLI, Copilot, Windsurf, Cline, Aider, Goose, OpenCode
brainstorm-ideas → identify-assumptions → brainstorm-experiments → pre-mortem
↓ ↓
create-prd → brainstorm-okrs → outcome-roadmap → prioritization-frameworks
↓
backlog-refinement → story-splitting → (build) → release-notes
↓
status-update-generator → (weekly exec comms)
python scrum-master/scripts/sprint_capacity_calculator.py team.json
python scrum-master/scripts/velocity_analyzer.py sprints.json
python execution/prioritization-frameworks/scripts/prioritization_scorer.py backlog.json --framework rice
python execution/cycle-time-analyzer/scripts/flow_metrics.py issues.jsonpython execution/dependency-map/scripts/dependency_graph.py deps.json --format mermaid
python execution/daci-framework/... # Document decision rights
python execution/status-update-generator/scripts/status_generator.py --format confluence| Metric | Target |
|---|---|
| Sprint Predictability | +40% improvement |
| Project On-Time Delivery | +25% improvement |
| Atlassian/Linear Operations Efficiency | +70% time savings |
| Documentation Findability | +60% improvement |
| Meeting Summary Quality | +50% consistency |
| Stakeholder Communication Coverage | 100% mapped |
| Weekly Status Update Time | 4 hr → 20 min |
| PM Onboarding Time | 90 days → 30 days |
- Shared Output Schema:
SHARED_OUTPUT_SCHEMA.md - Worked Examples:
examples/ - Implementation Summary:
IMPLEMENTATION_SUMMARY.md - Installation Guide:
INSTALLATION_GUIDE.txt - Real-World Scenario:
REAL_WORLD_SCENARIO.md - Main Documentation:
../CLAUDE.md
Last Updated: 2026-05-21 Skills Deployed: 54 production-ready PM skills Python Tools: 15+ CLI tools (standard library only) Integrations: Jira, Linear, Notion, Confluence, GitHub Projects, Atlassian MCP