All notable changes to the PM Toolkit plugin are documented here.
-
Hook script resolution —
hooks/hooks.jsonreferenced scripts via$CLAUDE_PROJECT_DIR, which expanded to the user's current project rather than the plugin install directory. All five hook commands (SessionStart,Stop,PostToolUse×2,PostToolUseFailure) now use${CLAUDE_PLUGIN_ROOT}, so hooks resolve correctly in every project. Previously every Claude Code session in an unrelated project produced non-blocking "No such file or directory" errors for all plugin hooks. -
Usage logger path —
hooks/usage-logger.shwrote.pm-toolkit-audit.jsonlinto each user's project directory. The audit log now lives at${XDG_STATE_HOME:-$HOME/.claude}/pm-toolkit/audit.jsonl(created automatically), keeping user repos clean. Workflow tracker state remains per-project by design; README notes adding.pm-toolkit-workflow-state.jsonto.gitignoreif undesired.
Added comprehensive migration planning capabilities for PE-backed SaaS companies. Covers platform rewrites, data migrations, product consolidations, vendor switches, and legacy sunsetting. Feature parity analysis is the central artifact — the document that determines whether migration is possible.
45 skills + 5 agents + 5 workflows
-
pm-migration-planner(613 lines) — The most comprehensive skill in the toolkit. 8-phase migration planning with 5 migration types (Platform, Data, Product Consolidation, Vendor/API, Compliance). Includes: strategy selection framework (Big Bang / Strangler Fig / Parallel Run / Cohort-Based / Feature Flag), feature parity matrix with customer-weighted gap analysis, data migration plan with 3 validation gates, 5-gate go/no-go framework with named decision-makers, 8-touch customer communication playbook, post-migration validation checklist, and legacy sunset plan. -
pm-pe-migration-report— PE board migration progress dashboard. One-page format PE operating partners can digest in 5 minutes: ARR scorecard (migrated/remaining/at-risk/lost), wave status, synergy realization tracker, cost vs budget, top 5 risks, post-migration customer health. -
pm-workflow-migration— 5-stage migration workflow: Migration PRD → Migration Planner → Risk Register → Stakeholder Simulation → Comms + PE Report. Multi-session with mid-entry support. Stages 1-3 produce the 100-day plan; Stages 4-5 repeat per wave. -
PE/Acquisition Context section in
domain-context.md— PE relationship, acquired companies table, target platform, migration constraints.
-
pm-risk-register— Added migration risk mode with 7 migration-specific categories: Data Integrity, Feature Parity, Customer Impact, Rollback, Timeline, Capacity, Integration. -
pm-customer-success— Added Stage G: Migration Support with health indicators (prep completion, support tickets, first login, workflow completion, data reconciliation, CSAT), intervention triggers, and CSM script templates. -
pm-stakeholder-simulator— Added 4 PE-specific stakeholder archetypes: PE Operating Partner, Acquired Company GM, Acquired Company PM, Acquired Company's Largest Customer. -
pm-hub— Added pm-migration-planner to Strategy, pm-pe-migration-report to Communication, pm-workflow-migration to Workflows. -
pm-ops-facilitator— Added migration routing.
Comprehensive quality pass driven by 4-persona testing (Junior PM, Senior PM, Head of Product, VP Product). Added personal context, fixed all critical gaps, refreshed older skills, and prepared for public distribution.
-
personal-context.md— Personal profile template for role, seniority, communication preferences, skill self-assessment (1-5), and "what to avoid" instructions. Skills adapt depth and tone to the user's experience level. Ships aspersonal-context.example.md(template in repo);personal-context.mdis gitignored for privacy. -
Mid-entry support in all 3 workflows — PMs with partial work can now skip stages:
pm-workflow-problem-to-prd: Enter at any of 5 stages with existing artifactspm-workflow-sales-enablement: Enter at any of 4 stagespm-workflow-quarterly-cycle: Enter at any of 5 stages
-
Revenue impact modeling (Phase 7) in
pm-pricing— 3-scenario revenue sensitivity model, break-even analysis, and sensitivity checks. Answers "what does this mean in euros?" -
Price migration strategy (Phase 8) in
pm-pricing— Grandfather/sunset/immediate/hybrid options, customer communication plan, at-risk account playbook -
2 new objection categories in
pm-objection-handler:- Internal Politics & Champion Risk (organizational dynamics, not product evaluation)
- Scope & Unbundling (buyer needs only part of the product)
-
Mid-quarter review template in
pm-workflow-quarterly-cycle— OKR progress check, bet status, capacity reality check, and course-correction decisions -
Deal-stage artifact map in
pm-gtm-strategistagent — Maps which GTM skill and artifact the sales team needs at each deal stage (Prospecting through Closed Lost) -
Verdict roll-up rules in
pm-review— Explicit Red/Yellow/Green to verdict mapping + "After the Review" guidance for fixing issues and re-reviewing -
Getting Started beginner path in README — 5-step guided journey for new users
-
Frameworks & Abbreviations glossary in README — 30+ terms explained (WTP, JTBD, SCQA, OKR, NRR, ARPU, MoSCoW, etc.) with "Used In" references
-
Skill Disambiguation table in README — Clarifies PRD vs brief vs one-pager, pricing vs AI pricing, OKR vs quarterly planning
-
Key Books Referenced table in README — Dunford, Ramanujam/Tacke, Cagan, Torres, Doerr
-
LICENSE (MIT), CONTRIBUTING.md, CLAUDE.md — Standard open-source files for public distribution
-
personal-context.md integrated across 21 files — 10 key skills, 5 agents, 3 workflows, README, hub. Each adapts guidance depth to the user's stated seniority and preferences.
-
6 older skills refreshed to v1.1+ quality — pm-swot, pm-board-deck, pm-meeting-notes, pm-executive-update, pm-one-pager, pm-stakeholder-simulator now have: Intent Detection, Pre-Population from Existing Artifacts, Anti-Patterns, and personal-context.md references
-
Hardcoded ERP examples removed from skills — Replaced with
domain-context.mdreferences so non-ERP companies get clean output -
domain-context.md Sales & GTM section — Populated with example data (previously empty placeholders). Consistent with the product sections that already shipped populated.
-
README overhauled — Install command updated to marketplace install, setup includes personal context, workflows section notes mid-entry support
.gitignoreupdated — personal-context.md, .pm-toolkit-audit.jsonl, .pm-toolkit-workflow-state.jsonpersonal-context.example.mdships as template; actual personal-context.md is user-local- Version bumped to 1.3.0 in plugin.json and marketplace.json
Sharpened 5 existing skills with proven product methodology from April Dunford ("Obviously Awesome") and Ramanujam & Tacke ("Monetizing Innovation"). Added a dedicated AI Pricing skill informed by Simon-Kucher, BVP, and Monetizely research.
43 skills + 5 agents + 4 workflows
pm-ai-pricing— New standalone skill for AI feature pricing and packaging. 9-phase process covering: embed-vs-charge decision framework, 6 AI pricing models (per-seat, per-agent, outcome-based, usage-based, hybrid), AI cost modeling with margin analysis and scaling scenarios, AI value measurement ("Worth 10x" test), AI credits and packaging strategies, AI-specific WTP research ("copilot tax"), AI pricing metrics (attach rate, AI NRR, cost-to-serve), and 7 AI pricing anti-patterns.
-
pm-pricing— Added 4 new sections from Ramanujam & Tacke's "Monetizing Innovation":- Phase 1B: Monetization Risk Diagnostic — diagnose Feature Shock, Minivation, Hidden Gem, and Undead failure patterns before pricing
- Phase 4B: Gabor-Granger Price Point Testing — complements existing Van Westendorp with demand curve and revenue-maximizing price analysis
- Phase 5B: Behavioral Pricing Principles — anchoring, compromise effect, decoy pricing, pennies-a-day framing, bundling vs unbundling
- Phase 5C: WTP-Based Segmentation — segment customers by willingness-to-pay, not demographics; map tiers to WTP clusters
-
pm-messaging-framework— Replaced generic positioning statement template with April Dunford's 5-component positioning framework:- Competitive Alternatives (what customers use today — not just competitors)
- Unique Attributes (what only you have)
- Value mapped to attributes (your unique features enable your unique value)
- Target Customer Characteristics (who cares a lot)
- Market Category (Head-to-Head / Big Fish Small Pond / Create New Category)
-
pm-battlecard— Added Dunford's competitive alternatives concept:- Reframed Question 1 from "Which competitor?" to "What is the prospect using today?" (includes manual processes, consultants, status quo)
- Added Section 1B "Alternative Type" with selling strategies for 5 alternative types
- Added status-quo-specific landmine questions for selling against manual processes
-
pm-sales-deck— Grounded in Dunford positioning:- Added Positioning Foundation Check before slide generation (competitive alternative, unique value, market category)
- Modified Slide 5 (Solution Intro) with market category framing
- Modified Slide 7 (Differentiation) to ground in buyer's actual competitive alternative
-
pm-value-prop-canvas— Added Positioning Bridge section mapping VPC findings to Dunford's 5 positioning components, with natural handoff to/pm-messaging-framework
- April Dunford, Obviously Awesome: How to Nail Product Positioning So Customers Get It, Buy It, Love It
- Madhavan Ramanujam & Georg Tacke, Monetizing Innovation: How Smart Companies Design the Product Around the Price (Simon-Kucher)
- Simon-Kucher, "Best Practices for Generative AI Packaging and Pricing"
- Bessemer Venture Partners, "The AI Pricing and Monetization Playbook"
- Monetizely, "The 2026 Guide to SaaS AI and Agentic Pricing Models"
- BCG, "Winning Strategies of Hypergrowth SaaS Champions"
Added a complete Go-to-Market category with 6 new sales-facing skills, a Sales Enablement workflow, and a GTM Strategist agent. Also added standalone PRD and OKR skills that were previously locked inside multi-stage workflows. Removed pm-api-docs (developer concern, not PM).
42 skills + 5 agents + 4 workflows
-
pm-prd— Standalone PRD generation with 4 modes: Full PRD, Lightweight Feature Spec, Iteration/V2 Spec, Migration/Compliance Spec. Built-in quality checks enforce problem-first thinking, leading+lagging metrics, scope discipline, and compliance awareness. Accepts prior discovery artifacts (JTBD, personas, opportunity trees) to pre-populate sections. -
pm-okr— Full OKR lifecycle skill with 5 modes: Define (create new OKRs), Refine (critique existing), Check-in (mid-quarter review), Score (quarter-end grading with classic/binary/traffic-light scoring), Align (map OKRs across company/team/individual levels). Enforces outcome-over-output writing, 60-70% confidence targeting, and leading+lagging KRs. -
Go-to-Market category (6 new skills):
pm-messaging-framework— Positioning statements, value pillars, elevator pitches (10s/30s/2min), per-persona and per-competitor messaging, proof points librarypm-battlecard— Per-competitor 1-page sales battlecards with feature comparison, objection responses, landmine questions, pricing comparison, deal-stage guidancepm-objection-handler— Objection playbook organized by category (price, product gaps, competitor claims, risk, timing, integration) with Acknowledge-Reframe-Evidence-Redirect frameworkpm-sales-deck— Sales/pitch deck content with 12-slide B2B structure, slide-by-slide talk tracks, and anticipated questionspm-gtm-launch— Go-to-market launch playbook with P0/P1/P2 tier scaling, cross-functional checklists, enablement timelines, communication plans, success metricspm-win-loss— Win/loss analysis with two modes: interview guide generation and data pattern analysis
-
pm-workflow-sales-enablement— 4-stage workflow: Messaging Framework → Battlecards → Objection Playbook → GTM Launch Playbook. All artifacts written for revenue teams, not product teams. -
pm-gtm-strategistagent — Routes to all 6 GTM skills. Bridges product strategy and revenue execution. -
Sales & GTM Context section in
domain-context.md— ICP definition, deal stages, sales team structure, buyer personas (distinct from product user personas), win/loss patterns
-
pm-workflow-problem-to-prd— Stage 4 now delegates topm-prd(Mode A) instead of maintaining an inline PRD template. Single source of truth for PRD generation. -
pm-workflow-quarterly-cycle— Stage 3 now delegates topm-okr(Mode A) instead of maintaining an inline OKR template. Single source of truth for OKR generation. -
pm-hub— Added Go-to-Market category, pm-prd in Foundation, pm-okr in Strategy, routing disambiguations (standalone PRD vs workflow, standalone OKR vs quarterly cycle)
pm-api-docs— Developer documentation tool (OpenAPI/Swagger, endpoint docs). Not a PM concern. Removed from skill set, hub, and README.
The PM Toolkit — a Claude Code plugin for product leadership at B2B SaaS companies.
35 skills + 4 agents + 3 workflows
Foundation: pm-hub, pm-review, pm-pricing, pm-north-star, pm-metric-framework, pm-onboarding-flow, pm-release-notes
Discovery & Data: pm-jtbd, pm-persona-generator, pm-feature-requests, pm-funnel-analyzer, pm-ab-test, pm-feedback-categorizer
Strategy: pm-quarterly-planning, pm-strategic-review, pm-swot, pm-value-prop-canvas, pm-opportunity-tree, pm-devils-advocate, pm-build-vs-buy, pm-risk-register
Communication: pm-executive-update, pm-decision-document, pm-stakeholder-simulator, pm-product-brief, pm-one-pager, pm-board-deck, pm-meeting-notes
Operations: pm-sprint-retro, pm-postmortem, pm-process-docs, pm-customer-success, pm-dashboard-designer, pm-journey-map, pm-api-docs
- pm-workflow-problem-to-prd — Problem → JTBD → Persona → Opportunity Tree → PRD → Review
- pm-workflow-competitive-intel — SWOT → Market Sizing → Competitive Profile → Positioning → One-Pager
- pm-workflow-quarterly-cycle — Strategic Review → North Star → OKRs → Planning → Roadmap
- pm-discovery-lead — Orchestrates discovery work through 4 phases
- pm-data-analyst — Orchestrates data analysis and metrics
- pm-ops-facilitator — Orchestrates PM operations and ceremonies
- pm-comms-writer — Orchestrates PM communication with SCQA framework
Every skill enforces: (1) Problem clarity before solutions, (2) Leading + lagging indicators for every metric, (3) Explicit non-goals and scope discipline, (4) Domain awareness via domain-context.md, (5) Hybrid interaction — ask critical questions, generate draft, iterate.
All domain-specific knowledge lives in domain-context.md. Ships with a B2B SaaS ERP example (German SMB market, DACH region, DATEV ecosystem). Replace with your own company context.