pm-claude-skills 43.0.0 → 45.0.0

package/README.md

@@ -1,4 +1,4 @@
-# 🧠 PM Skills — 451 Professional Agent Skills for Claude, ChatGPT, Gemini, Cursor, Codex & Hermes
+# 🧠 PM Skills — 454 Professional Agent Skills for Claude, ChatGPT, Gemini, Cursor, Codex & Hermes
 
 [![In the official Anthropic plugin directory](https://img.shields.io/badge/Anthropic%20Plugin%20Directory-Published-D97757?logo=anthropic&logoColor=white)](#-quick-install-2-minutes)
 [![Stars](https://img.shields.io/github/stars/mohitagw15856/pm-claude-skills?style=social)](https://github.com/mohitagw15856/pm-claude-skills/stargazers)
@@ -9,7 +9,8 @@
 [![MCP](https://img.shields.io/badge/MCP-registry-0098FF?logo=modelcontextprotocol&logoColor=white)](https://github.com/mohitagw15856/pm-claude-skills#-use-it-anywhere--the-ai-ecosystem)
 [![Remote MCP](https://img.shields.io/badge/ChatGPT%20%C2%B7%20Claude.ai-connector-10A37F)](mcp-remote/)
 [![Smithery](https://smithery.ai/badge/mohit15856/pm-skills)](https://smithery.ai/servers/mohit15856/pm-skills)
-[![Skills](https://img.shields.io/badge/skills-451-blue)](https://github.com/mohitagw15856/pm-claude-skills)
+[![Skills](https://img.shields.io/badge/skills-454-blue)](https://github.com/mohitagw15856/pm-claude-skills)
+[![Free runs served](https://img.shields.io/endpoint?url=https%3A%2F%2Fpm-skills-mcp.pm-claude-skills.workers.dev%2Ftry%2Fstats&cacheSeconds=300)](https://mohitagw15856.github.io/pm-claude-skills/)
 [![eval](https://img.shields.io/badge/eval--scored-208%20%C2%B7%20avg%204.8%2F5-brightgreen)](https://mohitagw15856.github.io/pm-claude-skills/leaderboard.html)
 [![Subagents](https://img.shields.io/badge/subagents-4-blueviolet)](agents/)
 [![Commands](https://img.shields.io/badge/slash%20commands-21-blueviolet)](commands/)
@@ -25,7 +26,7 @@
 
 <p align="center">
   <a href="https://mohitagw15856.github.io/pm-claude-skills/">
-    <img src="https://readme-typing-svg.demolab.com?font=Inter&weight=600&size=24&pause=900&color=D97757&center=true&vCenter=true&width=720&lines=451+pro+AI+skills;PRDs%2C+roadmaps%2C+launches+%E2%80%94+done+right;Run+any+one+free+in+your+browser" alt="PM Skills" />
+    <img src="https://readme-typing-svg.demolab.com?font=Inter&weight=600&size=24&pause=900&color=D97757&center=true&vCenter=true&width=720&lines=454+pro+AI+skills;PRDs%2C+roadmaps%2C+launches+%E2%80%94+done+right;Run+any+one+free+in+your+browser" alt="PM Skills" />
   </a>
 </p>
 
@@ -74,11 +75,11 @@
 </details>
 
 > **PM stands for Professional, not just Product Management.**
-> 451 professional skills + 4 agent templates across 65 bundles covering 25 professions. Built for Claude Code — and now portable to ChatGPT, Gemini, and Hermes Agent. Built by a PM, used by everyone.
+> 454 professional skills + 4 agent templates across 65 bundles covering 25 professions. Built for Claude Code — and now portable to ChatGPT, Gemini, and Hermes Agent. Built by a PM, used by everyone.
 
 A community-built library of professional skills for every field — product management, engineering, customer success, marketing, social media, writers, design, legal, finance, HR, sales, operations, research, and more. Each skill is a structured `SKILL.md` file that teaches an AI assistant how to produce professional-grade outputs for your workflows. Skills run natively in **Claude Code** and **Hermes Agent** (same open `SKILL.md` standard), and ship as ready-to-paste exports for **ChatGPT** and **Gemini** — see [Works With](#-works-with--cross-tool-compatibility).
 
-**🆕 v43.0.0 — the local-first & learning wave:** 🗂 connect the [Firm](https://mohitagw15856.github.io/pm-claude-skills/firm.html) & [Boardroom](https://mohitagw15856.github.io/pm-claude-skills/boardroom.html) to a **real folder on disk** (minutes, verdicts & predictions land as markdown Claude Code can read), 🔗 share a whole Boardroom session as a **replay link** (the URL *is* the data — no server), 🪑 swappable **bench presets** (🦈 fundraising / 🏥 healthcare / 🚀 seed-stage) + build-your-own executives + [bench packs in the registry](community/), 🎤 **[The Panel](https://mohitagw15856.github.io/pm-claude-skills/hiring.html)** — YOU interview an AI candidate with a hidden archetype and a coach scores your questioning, 🎓 **[The Academy](https://mohitagw15856.github.io/pm-claude-skills/academy.html)** — 18 anti-pattern drills across 3 tracks with live arena exams (no key needed), 🎁 **[Wrapped](https://mohitagw15856.github.io/pm-claude-skills/wrapped.html)** — your practice as an animated story + share card, 🩺 `npx pm-claude-skills doctor`, 🧬 a blind-judged [evolution loop](scripts/evolve-skills.mjs), and [Teams as a folder convention](docs/TEAMS-DECISION.md). Plus the gap-closing skills wave: **🧭 [`pm-method`](plugins/pm-method)** (8 process-discipline skills: `interview-me`, `brainstorming`, `writing-plans`/`executing-plans`, `verification-before-completion`…), real **Office file production** (stdlib `xlsx`/`docx`/`pptx` tools — no pip installs), and design craft (`frontend-design`, `brand-guidelines`). *Earlier — v42:* depth (`references/` + `templates/`) on all Production-Ready skills, 🔊 Theatre voices, Firm calibration & the CI smoke suite. *Earlier — v41, the modern skills wave:* two new domains — **🤖 [`pm-agentnative`](plugins/pm-agentnative)** (build for non-human users: [`mcp-server-spec`](skills/mcp-server-spec/SKILL.md), [`agent-readiness-audit`](skills/agent-readiness-audit/SKILL.md), [`agent-era-pricing`](skills/agent-era-pricing/SKILL.md), [`human-in-the-loop-design`](skills/human-in-the-loop-design/SKILL.md), [`voice-agent-design`](skills/voice-agent-design/SKILL.md)) and **🧑‍💼 [`pm-aiwork`](plugins/pm-aiwork)** (the questions every manager has this year: [`ai-roi-audit`](skills/ai-roi-audit/SKILL.md), [`role-redesign-for-ai`](skills/role-redesign-for-ai/SKILL.md), [`ai-usage-policy`](skills/ai-usage-policy/SKILL.md), [`ai-assisted-performance-review`](skills/ai-assisted-performance-review/SKILL.md), [`ai-content-audit`](skills/ai-content-audit/SKILL.md)) — plus 5 singles ([`ai-code-review`](skills/ai-code-review/SKILL.md), [`synthetic-user-research`](skills/synthetic-user-research/SKILL.md), [`async-decision-memo`](skills/async-decision-memo/SKILL.md), [`brand-impersonation-response`](skills/brand-impersonation-response/SKILL.md), [`feature-sunset-plan`](skills/feature-sunset-plan/SKILL.md)). Also: 🌠 Galaxy shooting stars + a 🎲 Warp button, 🎓 frameable [certificates](web/export-doc.js) from the arena pages, and 🔌 [MCP pairings](connectors/mcp-pairings.md) — skills that act through your connected servers. *Earlier — v40: [The Firm](https://mohitagw15856.github.io/pm-claude-skills/firm.html). v39 (three waves):* ⚡ *Superpowers:* [SkillBench](skillbench/) (the professional-work benchmark for models) · the [community registry](community/) (npm-for-skills) · the [Interview Gauntlet](https://mohitagw15856.github.io/pm-claude-skills/gauntlet.html) · [CI recipes](action/examples/) + [ambient hooks](hooks/) · Hugging Face [dataset publishing](dataset/) · [localization to 8 languages](i18n/). 🤯 *Breathtaking:* [living artifacts](https://mohitagw15856.github.io/pm-claude-skills/), [Document X-ray](https://mohitagw15856.github.io/pm-claude-skills/xray.html), [The Gym](https://mohitagw15856.github.io/pm-claude-skills/gym.html), [`pm-vision`](plugins/pm-vision), [`style-fingerprint`](skills/style-fingerprint/SKILL.md), [`evidence-lock`](skills/evidence-lock/SKILL.md). ✨ *Revamped:* the playground command bar + Galaxy 2.0 with a guided Sky tour. **451 skills · 65 bundles · 25 professions.** → [CHANGELOG](CHANGELOG.md) · [all releases](https://github.com/mohitagw15856/pm-claude-skills/releases).
+**🆕 v45.0.0 — the spectacle release:** ⚔️ **Duels** — challenge a friend to the *identical* Gym negotiation or Panel candidate (same hidden state, side-by-side scorecard), 🎓 **[The Charter](https://mohitagw15856.github.io/pm-claude-skills/charter.html)** — the performance-based, cryptographically verifiable certification, and **the library in 3D**: 🌌 [Galaxy 3D](https://mohitagw15856.github.io/pm-claude-skills/galaxy3d.html) (fly a bloom-lit starfield where your used skills burn brighter), 🏗 **[the Tower of Claims](https://mohitagw15856.github.io/pm-claude-skills/tower.html)** (your document as a physics tower — data is steel, assumptions are glass, watch it stand or **collapse**), 🏛 [the Stage](https://mohitagw15856.github.io/pm-claude-skills/stage.html) (Boardroom replays as cinema with a cracking hologram), ✨ holo persona cards, 🏆 [the Trophy Forge](https://mohitagw15856.github.io/pm-claude-skills/trophy.html), 🌃 [Skill City](https://mohitagw15856.github.io/pm-claude-skills/city.html), ☄️ the Reckoning orrery. Plus the record going public: **first official [SkillBench results](skillbench/results.json)** (Haiku 4.5 vs Sonnet 4.6), evolution loop run #1 (all champions held under blind judging), 3 sha256-pinned [registry packs](packs/), and a live **free-runs-served badge**. *Earlier — v44, the legendary release:* Campaign mode, the Professional Work Handbook (2,237-rule Anti-Pattern Almanac), the Reckoning, the ecosystem package manager, zero-key MCP sampling, pm-skills-tools, the SkillSpec badge & the census. *Earlier — v43, the local-first & learning wave:* the 🗂 workspace bridge, Boardroom replay links & bench packs, 🎤 The Panel, 🎓 The Academy, 🎁 Wrapped, the evolution loop & pm-method. *Earlier — v42:* depth (`references/` + `templates/`) on all Production-Ready skills, 🔊 Theatre voices, Firm calibration & the CI smoke suite. *Earlier — v41, the modern skills wave:* two new domains — **🤖 [`pm-agentnative`](plugins/pm-agentnative)** (build for non-human users: [`mcp-server-spec`](skills/mcp-server-spec/SKILL.md), [`agent-readiness-audit`](skills/agent-readiness-audit/SKILL.md), [`agent-era-pricing`](skills/agent-era-pricing/SKILL.md), [`human-in-the-loop-design`](skills/human-in-the-loop-design/SKILL.md), [`voice-agent-design`](skills/voice-agent-design/SKILL.md)) and **🧑‍💼 [`pm-aiwork`](plugins/pm-aiwork)** (the questions every manager has this year: [`ai-roi-audit`](skills/ai-roi-audit/SKILL.md), [`role-redesign-for-ai`](skills/role-redesign-for-ai/SKILL.md), [`ai-usage-policy`](skills/ai-usage-policy/SKILL.md), [`ai-assisted-performance-review`](skills/ai-assisted-performance-review/SKILL.md), [`ai-content-audit`](skills/ai-content-audit/SKILL.md)) — plus 5 singles ([`ai-code-review`](skills/ai-code-review/SKILL.md), [`synthetic-user-research`](skills/synthetic-user-research/SKILL.md), [`async-decision-memo`](skills/async-decision-memo/SKILL.md), [`brand-impersonation-response`](skills/brand-impersonation-response/SKILL.md), [`feature-sunset-plan`](skills/feature-sunset-plan/SKILL.md)). Also: 🌠 Galaxy shooting stars + a 🎲 Warp button, 🎓 frameable [certificates](web/export-doc.js) from the arena pages, and 🔌 [MCP pairings](connectors/mcp-pairings.md) — skills that act through your connected servers. *Earlier — v40: [The Firm](https://mohitagw15856.github.io/pm-claude-skills/firm.html). v39 (three waves):* ⚡ *Superpowers:* [SkillBench](skillbench/) (the professional-work benchmark for models) · the [community registry](community/) (npm-for-skills) · the [Interview Gauntlet](https://mohitagw15856.github.io/pm-claude-skills/gauntlet.html) · [CI recipes](action/examples/) + [ambient hooks](hooks/) · Hugging Face [dataset publishing](dataset/) · [localization to 8 languages](i18n/). 🤯 *Breathtaking:* [living artifacts](https://mohitagw15856.github.io/pm-claude-skills/), [Document X-ray](https://mohitagw15856.github.io/pm-claude-skills/xray.html), [The Gym](https://mohitagw15856.github.io/pm-claude-skills/gym.html), [`pm-vision`](plugins/pm-vision), [`style-fingerprint`](skills/style-fingerprint/SKILL.md), [`evidence-lock`](skills/evidence-lock/SKILL.md). ✨ *Revamped:* the playground command bar + Galaxy 2.0 with a guided Sky tour. **454 skills · 65 bundles · 25 professions.** → [CHANGELOG](CHANGELOG.md) · [all releases](https://github.com/mohitagw15856/pm-claude-skills/releases).
 
 ### ▶ See it in action — [try the live Skill Playground](https://mohitagw15856.github.io/pm-claude-skills/)
 
@@ -134,7 +135,7 @@ In Claude Code, [`/firm`](commands/firm.md) runs the same session natively — g
   <tr>
     <td width="50%">
       <a href="https://mohitagw15856.github.io/pm-claude-skills/"><img src="web/docs-assets/showcase/cmdbar.png" alt="The command bar — describe your task in plain words and get matched to the right skill instantly" /></a>
-      <b>🪄 <a href="https://mohitagw15856.github.io/pm-claude-skills/">The command bar</a></b> — don't browse 451 cards; just say it: <i>"a blameless postmortem for Friday's outage"</i> → the top three skills, instantly, locally, before you've even entered a key. Enter opens the best match.
+      <b>🪄 <a href="https://mohitagw15856.github.io/pm-claude-skills/">The command bar</a></b> — don't browse 454 cards; just say it: <i>"a blameless postmortem for Friday's outage"</i> → the top three skills, instantly, locally, before you've even entered a key. Enter opens the best match.
     </td>
     <td width="50%">
       <a href="https://mohitagw15856.github.io/pm-claude-skills/galaxy.html"><img src="web/docs-assets/showcase/galaxy.png" alt="Skill Galaxy — the library as a living star map with a guided sky tour" /></a>
@@ -228,7 +229,7 @@ Generic AI forgets everything between sessions, so you re-paste context forever
 <details>
 <summary>Read more ↓</summary>
 
-These 451 skills aren't a random catalog — they cover the **full arc of professional work**, end to end. Wherever you are in the loop, there's a skill for it:
+These 454 skills aren't a random catalog — they cover the **full arc of professional work**, end to end. Wherever you are in the loop, there's a skill for it:
 
 ```mermaid
 flowchart LR
@@ -349,6 +350,41 @@ Connect the Firm or the Boardroom to a **real folder on disk** (the 🗂 chip, C
 
 ---
 
+### 🌌 The library in 3D
+
+- **[Galaxy 3D](https://mohitagw15856.github.io/pm-claude-skills/galaxy3d.html)** — 454 skills as a bloom-lit starfield you fly through; **your sky is yours** (stars you've run burn brighter). **[Skill City](https://mohitagw15856.github.io/pm-claude-skills/city.html)** — the same idea as a dusk skyline: windows light up in buildings you've used.
+- **[The Tower of Claims](https://mohitagw15856.github.io/pm-claude-skills/tower.html)** — your document as a physics tower: data is steel, assumptions are glass, and the stress test shows the argument stand or **collapse**. Keyless demo built in.
+- **[The Stage](https://mohitagw15856.github.io/pm-claude-skills/stage.html)** — any Boardroom replay as a cinematic 3D scene: the hologram document cracks as objections land; the verdict seals it gold or shatters it.
+- **⚔️ Duels** — challenge links from the Gym and the Panel: your friend faces the *identical* hidden scenario, side-by-side scorecard at the end. **[The Charter](https://mohitagw15856.github.io/pm-claude-skills/charter.html)** — the performance-based, cryptographically verifiable certification. **[The Trophy Forge](https://mohitagw15856.github.io/pm-claude-skills/trophy.html)** — cast it in spinning gold.
+
+### 🗺 The Campaign, the Book, and the Ledger
+
+- **[Your First 90 Days](https://mohitagw15856.github.io/pm-claude-skills/campaign.html)** — campaign mode: one story across every arena. You join a fictional Series A startup, and your *real* arena scores write the plot — lose the Gym negotiation and the account churns; get fooled in the Panel and the bad hire haunts Week 10. Replayable, saves locally, certificate at Day 90.
+- **[The Professional Work Handbook](https://mohitagw15856.github.io/pm-claude-skills/handbook.html)** — the library as a book: 11 chapters of craft (the Production-Ready tier in full) + **the Anti-Pattern Almanac: 2,237 rules of professional judgment** from all 454 skills. Printable ([PDF](web/docs-assets/handbook.pdf)), free, regenerated every release.
+- **[The Reckoning](https://mohitagw15856.github.io/pm-claude-skills/reckoning.html)** — your prediction ledger: confidences, due dates, resurfacing, and a personal **calibration curve with a Brier score**. Also in the CLI: `npx pm-claude-skills reckoning`. When you say 80%, does it happen 80% of the time? Now you'll know.
+- **🔏 The chain of trust** — security scan → [sha256 content pinning](community/README.md#-trust--integrity--the-full-chain) in the registry → install lockfile + `npx pm-claude-skills verify` drift detection → npm provenance. Verifiable end to end, no trust in us required.
+
+### 🌐 Infrastructure for the whole ecosystem
+
+- **[`pm-skills-tools`](tools-pkg/)** — all 454 skills as **agent tools**: OpenAI function schemas, Vercel AI SDK adapters, `pick()`/`search()` runtime. `npm i pm-skills-tools` and your agent wields the library.
+- **[The SkillSpec badge](skillspec/README.md#the-badge)** — any skills repo can wear its live-graded conformance level: `![SkillSpec](https://img.shields.io/endpoint?url=…/badge?repo=you/yours)`.
+- **[The State of Agent Skills](docs/reports/)** — the census of the public SKILL.md ecosystem (counts, conformance pyramid, security hits), regenerated from GitHub code search. The observatory, not just a participant.
+- **[Every protocol](connectors/mcp-pairings.md#every-protocol-one-library)** — MCP (with zero-key sampling) · A2A · AGENTS.md · function calling · REST.
+- **[The Apprentice](templates/apprentice/)** — a ready-to-run Agent SDK agent grounded in your brain and your skills' Quality Checks. Fork it.
+- **`council`** (cross-vendor adversarial review) · **`migrate`** (convert prompt libraries into skills) · **[chief-of-staff hook](hooks/)** (every session opens with your predictions due and latest verdicts).
+
+### 📦 The ecosystem toolchain — beyond this library
+
+- **`npx pm-claude-skills install <owner/repo>`** — install skills from **any public GitHub repo**, security-scanned and [SkillSpec](SKILLSPEC.md)-graded on the way in; flagged skills are never installed. The package manager for the whole skills ecosystem, with auditing as the value add.
+- **`npx pm-claude-skills chain <workflow>`** — run a whole recipe headless (`chain --list`): each skill feeds the next, and `--deck`/`--doc` finish with a **real .pptx/.docx**. Raw notes in, board-ready deck out, one command.
+- **`npx pm-claude-skills init`** — the 60-second professional workspace: `brain/`, standing context, arena folders, CLAUDE.md wiring. Then `doctor` keeps it healthy.
+- **`run_skill` over MCP with zero API key** — the [MCP server](mcp/) executes skills via **MCP sampling**: the generation runs on your client's model, no key needed by anyone.
+- **[`skillspec-check`](skillspec/)** — the eslint for SKILL.md, as a standalone npm package + [pre-commit hook](.pre-commit-hooks.yaml): lint ANY repo's skills against the standard (`--min-level 2` as a CI gate).
+- **[Boardroom doc review in CI](action/examples/doc-review-boardroom.yml)** — PRs touching docs get a CFO/CTO/CCO critique with a 🟢/🟡/🔴 Chair verdict as a PR comment. Documents get code review.
+- **🧮 Skills that do math** — [`cohort-curve-model`](skills/cohort-curve-model/SKILL.md), [`runway-monte-carlo`](skills/runway-monte-carlo/SKILL.md), [`pricing-sensitivity-model`](skills/pricing-sensitivity-model/SKILL.md): each ships a stdlib script that **computes** (curve fitting, 5,000-path simulation, survey crossings) and emits a **real .xlsx with live formulas**.
+
+---
+
 ## 🧠 Skill Memory — set your context once, every skill uses it
 
 <details>
@@ -429,7 +465,7 @@ Most skill repos are a folder of prompts. This one is a **system** — measured,
 
 | | **PM Skills** | Typical skill repo |
 |---|:---:|:---:|
-| Skills | **451**, across 25 professions | a handful → dozens, usually one domain |
+| Skills | **454**, across 25 professions | a handful → dozens, usually one domain |
 | Quality | **eval-scored** on a rubric + a public [benchmark](https://mohitagw15856.github.io/pm-claude-skills/benchmark.html) | trust the README |
 | Improves itself | ✅ eval → critique → rewrite (kept only if it scores higher) | ✗ |
 | Grounded in frameworks | ✅ each cites its source (RICE, JTBD, Pyramid Principle…) | rarely |
@@ -907,7 +943,7 @@ More templates will follow. If you want to contribute one, see the [template con
 <details>
 <summary>Read more ↓</summary>
 
-**Latest: v43.0.0 — the local-first & learning wave: the workspace bridge to a real folder, Boardroom replay links & bench packs, The Panel (you're the interviewer), The Academy, Wrapped, `doctor`, the evolution loop, plus `pm-method` + real Office-file production (451 skills, 65 bundles)** · *v42:* depth on all Production-Ready skills, arena voices, Firm calibration & the web-tests safety net · *v41, modern skills:* [`pm-agentnative`](plugins/pm-agentnative) + [`pm-aiwork`](plugins/pm-aiwork) (15 skills for the agent era), Galaxy cosmic events, 🎓 certificates & [MCP pairings](connectors/mcp-pairings.md) · *v40: [The Firm](https://mohitagw15856.github.io/pm-claude-skills/firm.html) · v39, three waves:* [SkillBench](skillbench/), the [community registry](community/), the [Interview Gauntlet](https://mohitagw15856.github.io/pm-claude-skills/gauntlet.html), [living artifacts](https://mohitagw15856.github.io/pm-claude-skills/), [Document X-ray](https://mohitagw15856.github.io/pm-claude-skills/xray.html), [The Gym](https://mohitagw15856.github.io/pm-claude-skills/gym.html), pm-vision, i18n, HF publishing, CI recipes & the Galaxy Sky tour. 451 skills across 65 bundles, 25 professions. Every tagged release has complete notes on the **[Releases page](https://github.com/mohitagw15856/pm-claude-skills/releases)**.
+**Latest: v45.0.0 — the spectacle release: ⚔️ Duels, 🎓 the verifiable Charter, and the library in 3D (Galaxy 3D, the collapsing Tower of Claims, the Stage, holo cards, the Trophy Forge, Skill City, the orrery) — plus the first official SkillBench results and the first evolution-loop run (454 skills, 65 bundles)** · *v44:* Campaign mode, the Handbook, the Reckoning, the ecosystem package manager, zero-key MCP sampling, pm-skills-tools, the badge & the census · *v43:* the workspace bridge, replay links & bench packs, The Panel, The Academy, Wrapped, the evolution loop & pm-method · *v42:* depth on all Production-Ready skills, arena voices, Firm calibration & the web-tests safety net · *v41, modern skills:* [`pm-agentnative`](plugins/pm-agentnative) + [`pm-aiwork`](plugins/pm-aiwork) (15 skills for the agent era), Galaxy cosmic events, 🎓 certificates & [MCP pairings](connectors/mcp-pairings.md) · *v40: [The Firm](https://mohitagw15856.github.io/pm-claude-skills/firm.html) · v39, three waves:* [SkillBench](skillbench/), the [community registry](community/), the [Interview Gauntlet](https://mohitagw15856.github.io/pm-claude-skills/gauntlet.html), [living artifacts](https://mohitagw15856.github.io/pm-claude-skills/), [Document X-ray](https://mohitagw15856.github.io/pm-claude-skills/xray.html), [The Gym](https://mohitagw15856.github.io/pm-claude-skills/gym.html), pm-vision, i18n, HF publishing, CI recipes & the Galaxy Sky tour. 454 skills across 65 bundles, 25 professions. Every tagged release has complete notes on the **[Releases page](https://github.com/mohitagw15856/pm-claude-skills/releases)**.
 
 Full [Keep a Changelog](https://keepachangelog.com/)-format history — every release back to the start — is in **[CHANGELOG.md](CHANGELOG.md)**.
 
@@ -1018,7 +1054,7 @@ Every skill, grouped by profession. **[Browse the full per-skill catalog → SKI
 <details>
 <summary>Read more ↓</summary>
 
-Building and maintaining 451 skills across 65 bundles takes real time — testing skills against new model releases, building new ones from community requests, writing the article series, and keeping documentation current.
+Building and maintaining 454 skills across 65 bundles takes real time — testing skills against new model releases, building new ones from community requests, writing the article series, and keeping documentation current.
 
 If these skills save you time at work — or you're a company that wants your logo in front of the PMs, engineers, and operators who use them daily — **[become a sponsor →](https://github.com/sponsors/mohitagw15856)** (or [☕ buy me a coffee](https://www.buymeacoffee.com/mohit15856)).
 

package/bin/chain.mjs

@@ -0,0 +1,113 @@
+// `pm-claude-skills chain <workflow>` — run a whole workflow recipe headless:
+// every skill in the chain executes in order, each output carried forward as
+// context for the next. Raw notes in, a folder of finished artifacts out —
+// optionally including REAL Office files via the stdlib document tools.
+//
+//   pm-claude-skills chain --list
+//   pm-claude-skills chain run-discovery --input notes.txt
+//   pm-claude-skills chain ship-a-feature --text "referral program for B2B" --out-dir ./artifacts
+//   cat notes.md | pm-claude-skills chain close-the-quarter --deck   # + a real .pptx at the end
+//
+// Needs ANTHROPIC_API_KEY (your key, one call per step). Pure Node + optional
+// python3 only for --deck/--doc (the same zero-dependency scripts the document
+// skills ship).
+import { readFileSync, existsSync, writeFileSync, mkdirSync } from 'node:fs';
+import { join, dirname } from 'node:path';
+import { fileURLToPath } from 'node:url';
+import { spawnSync } from 'node:child_process';
+import { complete, parseSkill } from './lib/anthropic.mjs';
+
+const PKG_ROOT = dirname(dirname(fileURLToPath(import.meta.url)));
+const getArg = (argv, name, def) => { const i = argv.indexOf('--' + name); return i !== -1 && argv[i + 1] ? argv[i + 1] : def; };
+
+const SUFFIX =
+  '\n\n---\nExecute this skill now on the material below and produce the complete output. ' +
+  'Do not ask follow-up questions — work with what is given and note any reasonable assumptions.';
+
+async function readStdin() {
+  if (process.stdin.isTTY) return '';
+  const chunks = [];
+  for await (const c of process.stdin) chunks.push(c);
+  return Buffer.concat(chunks).toString('utf8').trim();
+}
+
+function loadWorkflows() {
+  try { return JSON.parse(readFileSync(join(PKG_ROOT, 'workflows.json'), 'utf8')).workflows || []; }
+  catch { return []; }
+}
+
+export async function run(argv) {
+  const workflows = loadWorkflows();
+  const id = argv[0] && !argv[0].startsWith('--') ? argv[0] : '';
+
+  if (argv.includes('--list') || (!id && !argv.includes('--help'))) {
+    console.log('Workflow chains — run several skills as one pipeline, each output feeding the next:\n');
+    for (const w of workflows) console.log(`  ${w.id.padEnd(22)} ${w.steps.map((s) => s.skill).join(' → ')}`);
+    console.log('\nRun one:  pm-claude-skills chain <id> --text "…" | --input <file> | stdin   [--out-dir <dir>] [--model <m>] [--deck] [--doc]');
+    console.log('--deck / --doc additionally produce a real .pptx / .docx from the final artifact (needs python3, no pip installs).');
+    return id ? 0 : (argv.includes('--list') ? 0 : 1);
+  }
+  if (argv.includes('--help') || argv.includes('-h')) return run(['--list']);
+
+  const w = workflows.find((x) => x.id === id);
+  if (!w) { console.error(`Unknown chain "${id}". See them all:  pm-claude-skills chain --list`); return 1; }
+
+  const apiKey = process.env.ANTHROPIC_API_KEY || '';
+  if (!apiKey) { console.error('Set ANTHROPIC_API_KEY to run a chain (one API call per step).'); return 1; }
+
+  let input = getArg(argv, 'text', '');
+  const inFile = getArg(argv, 'input');
+  if (!input && inFile) {
+    if (!existsSync(inFile)) { console.error(`Input file not found: ${inFile}`); return 1; }
+    input = readFileSync(inFile, 'utf8');
+  }
+  if (!input) input = await readStdin();
+  if (!input.trim()) { console.error('No input. Provide --text "…", --input <file>, or pipe via stdin.'); return 1; }
+
+  const model = getArg(argv, 'model', 'claude-sonnet-4-6');
+  const outDir = getArg(argv, 'out-dir', `./${w.id}-artifacts`);
+  mkdirSync(outDir, { recursive: true });
+
+  console.error(`⛓  ${w.name} — ${w.steps.length} steps, model ${model}\n   ${w.steps.map((s) => s.skill).join(' → ')}\n`);
+  let carried = `THE ORIGINAL BRIEF / RAW MATERIAL:\n\n${input}`;
+  const outputs = [];
+
+  for (let i = 0; i < w.steps.length; i++) {
+    const step = w.steps[i];
+    const file = join(PKG_ROOT, 'skills', step.skill, 'SKILL.md');
+    if (!existsSync(file)) { console.error(`  ✗ step ${i + 1}: skill "${step.skill}" not found — skipping`); continue; }
+    const { body } = parseSkill(readFileSync(file, 'utf8'));
+    process.stderr.write(`  ${i + 1}/${w.steps.length} ${step.skill} … `);
+    const t0 = Date.now();
+    const out = await complete({ apiKey, model, system: body + SUFFIX, messages: [{ role: 'user', content: carried }], maxTokens: 8192 });
+    const fname = `${String(i + 1).padStart(2, '0')}-${step.skill}.md`;
+    writeFileSync(join(outDir, fname), out + '\n');
+    outputs.push({ skill: step.skill, file: fname, out });
+    console.error(`✔ ${((Date.now() - t0) / 1000).toFixed(1)}s → ${fname}`);
+    // Carry forward: the original material + this step's artifact (the chain's contract).
+    carried += `\n\n---\nOUTPUT OF THE PREVIOUS STEP (${step.skill} — ${step.produces}):\n\n${out}`;
+    if (carried.length > 150000) carried = carried.slice(-150000);   // context hygiene on long chains
+  }
+
+  if (!outputs.length) { console.error('\nNo steps ran.'); return 1; }
+  const last = outputs[outputs.length - 1];
+
+  // ── Optional: real Office files from the final artifact (stdlib python) ────
+  for (const [flag, script, ext, label] of [
+    ['--deck', join(PKG_ROOT, 'skills', 'slide-deck', 'scripts', 'pptx_tool.py'), 'pptx', 'deck'],
+    ['--doc', join(PKG_ROOT, 'skills', 'word-document', 'scripts', 'docx_tool.py'), 'docx', 'document'],
+  ]) {
+    if (!argv.includes(flag)) continue;
+    if (!existsSync(script)) { console.error(`  (${flag}: tool script missing — skipped)`); continue; }
+    const target = join(outDir, `${w.id}.${ext}`);
+    const args = ext === 'pptx'
+      ? [script, 'build', target, '--outline-file', join(outDir, last.file)]
+      : [script, 'create', target, '--text-file', join(outDir, last.file)];
+    const r = spawnSync('python3', args, { encoding: 'utf8' });
+    if (r.status === 0) console.error(`  📎 real ${label}: ${target}`);
+    else console.error(`  (${flag} failed: ${(r.stderr || r.stdout || '').split('\n')[0]})`);
+  }
+
+  console.error(`\n✅ ${outputs.length} artifacts in ${outDir}/ — final: ${last.file}`);
+  return 0;
+}

package/bin/cli.mjs

@@ -246,6 +246,13 @@ Usage:
   npx pm-claude-skills add --agent <claude|hermes|codex|openclaw|cursor|windsurf|aider> [--target <path>] [--link] [--dry-run]
   npx pm-claude-skills run <skill> [--text "…" | --input <file>] [--model <m>] [--out <file>]
   npx pm-claude-skills search [query…] [--json] [--limit <n>]
+  npx pm-claude-skills install <owner/repo>   # install skills from ANY GitHub repo — security-scanned + SkillSpec-graded
+  npx pm-claude-skills verify                 # integrity check: detect drift in anything "install" brought in
+  npx pm-claude-skills chain <workflow>       # run a whole multi-skill pipeline (chain --list to see them)
+  npx pm-claude-skills council <skill>        # author -> cross-vendor critique -> arbiter (2+ provider keys)
+  npx pm-claude-skills migrate <dir>          # convert a folder of prompts/SOPs into SkillSpec skills
+  npx pm-claude-skills init        # scaffold the professional workspace here (brain/, context, arena folders)
+  npx pm-claude-skills reckoning   # your prediction ledger: due calls, hit rate, calibration curve, Brier score
   npx pm-claude-skills doctor      # checkup: what's installed, what's stale, what to fix (read-only)
   npx pm-claude-skills list
   npx pm-claude-skills --version
@@ -261,6 +268,8 @@ Examples:
 
   npx pm-claude-skills run prd-template --text "a referral program for B2B users"   # run a skill (needs ANTHROPIC_API_KEY)
   cat notes.txt | npx pm-claude-skills run meeting-notes --out summary.md           # pipe input, write the artifact
+  cat notes.txt | npx pm-claude-skills chain run-discovery --deck                   # notes → 4 artifacts + a real .pptx
+  npx pm-claude-skills install acme/their-skills --dry-run                          # audit a third-party skill repo
   npx pm-claude-skills generate --from <url|file>   # turn your docs into a SKILL.md (needs ANTHROPIC_API_KEY)
 
 ${STAR}
@@ -269,10 +278,45 @@ ${STAR}
 const opts = parse(process.argv.slice(2));
 const cmd = opts._[0];
 if (opts.version) console.log(VERSION);
-else if (!cmd || cmd === 'help' || (opts.help && cmd !== 'run' && cmd !== 'generate')) console.log(HELP);
+else if (!cmd || cmd === 'help' || (opts.help && !['run', 'generate', 'install', 'chain', 'init', 'reckoning', 'council', 'migrate'].includes(cmd))) console.log(HELP);
 else if (cmd === 'list') list();
 else if (cmd === 'search') search(opts);
 else if (cmd === 'add') add(opts);
+else if (cmd === 'migrate') {
+  const { run } = await import('./migrate.mjs');
+  try { process.exit(await run(process.argv.slice(3))); }
+  catch (e) { console.error(`Error: ${e.message}`); process.exit(1); }
+}
+else if (cmd === 'council') {
+  const { run } = await import('./council.mjs');
+  try { process.exit(await run(process.argv.slice(3))); }
+  catch (e) { console.error(`Error: ${e.message}`); process.exit(1); }
+}
+else if (cmd === 'verify') {
+  const { run } = await import('./verify.mjs');
+  try { process.exit(await run(process.argv.slice(3))); }
+  catch (e) { console.error(`Error: ${e.message}`); process.exit(1); }
+}
+else if (cmd === 'reckoning') {
+  const { run } = await import('./reckoning.mjs');
+  try { process.exit(await run(process.argv.slice(3))); }
+  catch (e) { console.error(`Error: ${e.message}`); process.exit(1); }
+}
+else if (cmd === 'init') {
+  const { run } = await import('./init.mjs');
+  try { process.exit(await run(process.argv.slice(3))); }
+  catch (e) { console.error(`Error: ${e.message}`); process.exit(1); }
+}
+else if (cmd === 'chain') {
+  const { run } = await import('./chain.mjs');
+  try { process.exit(await run(process.argv.slice(3))); }
+  catch (e) { console.error(`Error: ${e.message}`); process.exit(1); }
+}
+else if (cmd === 'install') {
+  const { run } = await import('./install.mjs');
+  try { process.exit(await run(process.argv.slice(3))); }
+  catch (e) { console.error(`Error: ${e.message}`); process.exit(1); }
+}
 else if (cmd === 'doctor') {
   const { run } = await import('./doctor.mjs');
   try { process.exit(await run()); }

package/bin/council.mjs

@@ -0,0 +1,94 @@
+// `pm-claude-skills council <skill>` — the second-opinion machine. One model
+// AUTHORS the artifact, a DIFFERENT vendor's model CRITIQUES it against the
+// skill's own Quality Checks and Anti-Patterns, and a third (or the first)
+// ARBITRATES, producing the final artifact with the critique addressed.
+//
+// Cross-provider adversarial review: vendor blind spots don't overlap, so the
+// critic catches what the author's family reliably misses.
+//
+//   pm-claude-skills council prd-template --text "usage-based pricing for the API tier"
+//   pm-claude-skills council exec-summary --input notes.md --models anthropic,openai,gemini
+//   pm-claude-skills council prd-template --text "…" --dry-run     # plan only, no calls
+//
+// Keys from env: ANTHROPIC_API_KEY, OPENAI_API_KEY, GEMINI_API_KEY (needs ≥2
+// providers to be a council; 3 calls total). Output → stdout or --out <file>.
+import { readFileSync, existsSync, writeFileSync } from 'node:fs';
+import { join, dirname } from 'node:path';
+import { fileURLToPath } from 'node:url';
+import { complete, available, provider } from './lib/providers.mjs';
+
+const PKG_ROOT = dirname(dirname(fileURLToPath(import.meta.url)));
+const getArg = (argv, n, d) => { const i = argv.indexOf('--' + n); return i >= 0 && argv[i + 1] ? argv[i + 1] : d; };
+
+async function readStdin() {
+  if (process.stdin.isTTY) return '';
+  const chunks = []; for await (const c of process.stdin) chunks.push(c);
+  return Buffer.concat(chunks).toString('utf8').trim();
+}
+
+export async function run(argv) {
+  const skill = argv[0] && !argv[0].startsWith('--') ? argv[0] : '';
+  if (!skill || argv.includes('--help')) {
+    console.log(`The Council — cross-provider adversarial review. Author → critic (different vendor) → arbiter.
+
+Usage: pm-claude-skills council <skill> [--text "…" | --input <file> | stdin] [--models a,b,c] [--out <file>] [--dry-run]
+Providers by env key: ANTHROPIC_API_KEY, OPENAI_API_KEY, GEMINI_API_KEY (≥2 required; 3 API calls total).`);
+    return skill ? 0 : 1;
+  }
+  const file = join(PKG_ROOT, 'skills', skill, 'SKILL.md');
+  if (!existsSync(file)) { console.error(`Unknown skill "${skill}" — see: pm-claude-skills list`); return 1; }
+  const text = readFileSync(file, 'utf8');
+  const body = (text.match(/^---\r?\n[\s\S]*?\r?\n---\r?\n?([\s\S]*)$/) || [, text])[1].trim();
+  const qc = (body.match(/##\s*Quality Checks\s*\n([\s\S]*?)(?=\n##\s|$)/i) || [, '(none declared)'])[1].trim();
+  const ap = (body.match(/##\s*Anti-Patterns\s*\n([\s\S]*?)(?=\n##\s|$)/i) || [, '(none declared)'])[1].trim();
+
+  let input = getArg(argv, 'text', '');
+  const inFile = getArg(argv, 'input');
+  if (!input && inFile && existsSync(inFile)) input = readFileSync(inFile, 'utf8');
+  if (!input) input = await readStdin();
+  if (!input.trim()) { console.error('No input. Provide --text, --input <file>, or pipe via stdin.'); return 1; }
+
+  const wanted = (getArg(argv, 'models', '') || '').split(',').map((s) => s.trim()).filter(Boolean);
+  const avail = available();
+  const roster = wanted.length ? wanted : avail;
+  const dryRun = argv.includes('--dry-run');
+
+  if (roster.length < 2) {
+    console.error(`The Council needs ≥2 providers; ${avail.length ? 'available: ' + avail.join(', ') : 'no provider keys set'}.` +
+      ' Set at least two of ANTHROPIC_API_KEY / OPENAI_API_KEY / GEMINI_API_KEY.' + (dryRun ? '' : ' (--dry-run works without keys.)'));
+    if (!dryRun) return 1;
+  }
+  const [author, critic, arbiter] = [roster[0] || 'anthropic', roster[1] || 'openai', roster[2] || roster[0] || 'anthropic'];
+  console.error(`⚖️  The Council on "${skill}" — author: ${author} · critic: ${critic} · arbiter: ${arbiter}`);
+
+  if (dryRun) {
+    console.error(`[dry-run] 3 calls would run:
+  1. ${author} (${provider(author).defaultModel}) authors the artifact from the skill (${body.length} chars of instructions).
+  2. ${critic} (${provider(critic).defaultModel}) critiques it against the skill's OWN bar — Quality Checks (${qc.split('\n').length} items) + Anti-Patterns (${ap.split('\n').length} items) — verdict per check, quoting the draft.
+  3. ${arbiter} arbitrates: final artifact with every upheld objection addressed, and a changelog of what the critique changed.
+No API calls made.`);
+    return 0;
+  }
+
+  // 1. Author
+  console.error(`  1/3 ${author} authoring…`);
+  const draft = await complete(author, { system: body + '\n\n---\nExecute this skill now on the input and produce the complete artifact. No questions; state assumptions.', user: input, maxTokens: 8192 });
+
+  // 2. Cross-vendor critique against the skill's own bar
+  console.error(`  2/3 ${critic} critiquing…`);
+  const critique = await complete(critic, {
+    system: `You are the adversarial reviewer on a council. A DIFFERENT model produced a draft using a skill; judge the draft strictly against that skill's OWN declared bar — nothing else.\n\nTHE SKILL'S QUALITY CHECKS:\n${qc}\n\nTHE SKILL'S ANTI-PATTERNS:\n${ap}\n\nFor each quality check: PASS/FAIL + one line quoting the draft. For each anti-pattern: CLEAN/VIOLATED + where. End with "## Priority repairs" — the 3 fixes that most improve the artifact. Be specific; no praise padding.`,
+    user: `THE ORIGINAL INPUT:\n${input.slice(0, 8000)}\n\nTHE DRAFT UNDER REVIEW:\n${draft}`, maxTokens: 3000 });
+
+  // 3. Arbitration → final
+  console.error(`  3/3 ${arbiter} arbitrating…`);
+  const final = await complete(arbiter, {
+    system: body + '\n\n---\nYou are the arbiter of a council. Below: the original input, a draft, and a cross-examination of that draft against this skill\'s own quality bar. Produce the FINAL artifact: keep what survived, repair every upheld objection, ignore critiques that misread the input. Then append "## Council changelog" — 3-6 bullets on what the critique changed.',
+    user: `INPUT:\n${input.slice(0, 8000)}\n\nDRAFT (${author}):\n${draft}\n\nCRITIQUE (${critic}):\n${critique}`, maxTokens: 8192 });
+
+  const outFile = getArg(argv, 'out');
+  const full = final + `\n\n---\n*Council: ${author} authored · ${critic} critiqued · ${arbiter} arbitrated — pm-claude-skills*\n`;
+  if (outFile) { writeFileSync(outFile, full); console.error(`✅ ${outFile}`); }
+  else process.stdout.write(full);
+  return 0;
+}

package/bin/init.mjs

@@ -0,0 +1,111 @@
+// `pm-claude-skills init` — scaffold the 60-second professional workspace in the
+// current directory: the brain, your standing context, the arena folders the web
+// tools write into, and a CLAUDE.md section that wires it all together. Doctor
+// diagnoses; init prescribes and executes.
+//
+//   npx pm-claude-skills init             # scaffold (never overwrites anything)
+//   npx pm-claude-skills init --dry-run   # show what would be created
+//
+// Idempotent by design: every file is create-if-missing; existing files are
+// left untouched and reported. Pure Node, writes only inside the current dir.
+import { existsSync, mkdirSync, writeFileSync, readFileSync, appendFileSync, cpSync } from 'node:fs';
+import { join, dirname } from 'node:path';
+import { fileURLToPath } from 'node:url';
+
+const PKG_ROOT = dirname(dirname(fileURLToPath(import.meta.url)));
+const STAR = '⭐ Find this useful? Star the repo: https://github.com/mohitagw15856/pm-claude-skills';
+
+const CLAUDE_MD_SECTION = `
+## PM Skills — this project's professional setup
+
+- **Standing context**: read \`brain/context.md\` before professional-writing tasks (audience, voice, metrics). Honour the provenance tags in \`brain/\` — strongest to weakest: \`[data] [interview] [external] [verbal] [hunch]\`.
+- **Durable memory**: \`brain/\` holds decisions, hypotheses, stakeholders, and knowledge. After significant outcomes, append a dated line to the right file rather than letting the fact evaporate.
+- **Web arena artifacts**: \`firm-minutes/\` and \`boardroom/\` receive minutes, verdicts, and transcripts written by the pm-skills web tools (workspace bridge). Treat them as inputs — e.g. read the latest Boardroom verdict before revising the document it judged.
+- **Skills**: if the pm-skills library is installed (\`npx pm-claude-skills add --agent claude\`), prefer its skills for professional artifacts (PRDs, updates, postmortems…) over improvising structure.
+`;
+
+const AGENTS_MD = `# AGENTS.md
+
+Guidance for AI coding agents (Codex, Jules, and anything that reads this file) working in this repository.
+
+## Professional context
+
+- Read \`brain/context.md\` before drafting any professional artifact — it holds the company, audience, voice, and the metrics that matter. Honour the provenance tags in \`brain/\` (strongest → weakest): \`[data] [interview] [external] [verbal] [hunch]\`.
+- Durable memory lives in \`brain/\` (decisions, hypotheses, stakeholders, knowledge). Append dated lines after significant outcomes.
+
+## Skills
+
+- Professional skills (PRDs, updates, postmortems, models…) are available via the pm-claude-skills library. Discover and apply them instead of improvising document structure:
+  - CLI: \`npx pm-claude-skills search <task>\` then \`npx pm-claude-skills run <skill> --input <file>\`
+  - MCP: the \`pm-skills\` server exposes \`search_skills\` / \`get_skill\` / \`run_skill\`
+- A skill's Quality Checks and Anti-Patterns sections are the acceptance criteria for its output — verify against them before declaring done.
+
+## Artifacts
+
+- \`firm-minutes/\` and \`boardroom/\` contain minutes, verdicts, and transcripts written by the pm-skills web arenas. Treat them as inputs: read the latest Boardroom verdict before revising the document it judged.
+- Predictions live in \`brain/predictions/\` — settle due ones honestly (\`npx pm-claude-skills reckoning\`).
+`;
+
+export async function run(argv) {
+  const dryRun = argv.includes('--dry-run');
+  const cwd = process.cwd();
+  const made = [], kept = [];
+
+  const mk = (rel, content) => {
+    const p = join(cwd, rel);
+    if (existsSync(p)) { kept.push(rel); return; }
+    if (!dryRun) { mkdirSync(dirname(p), { recursive: true }); writeFileSync(p, content); }
+    made.push(rel);
+  };
+  const mkdir = (rel) => {
+    const p = join(cwd, rel);
+    if (existsSync(p)) { kept.push(rel + '/'); return; }
+    if (!dryRun) mkdirSync(p, { recursive: true });
+    made.push(rel + '/');
+  };
+
+  console.log(`${dryRun ? '[dry-run] ' : ''}Scaffolding the professional workspace in ${cwd}\n`);
+
+  // ── The brain: durable memory the skills read and write ────────────────────
+  const brainTpl = join(PKG_ROOT, 'templates', 'brain');
+  if (existsSync(brainTpl)) {
+    if (existsSync(join(cwd, 'brain'))) kept.push('brain/');
+    else { if (!dryRun) cpSync(brainTpl, join(cwd, 'brain'), { recursive: true }); made.push('brain/  (context, knowledge/, decisions/, hypotheses, stakeholders, entities — from templates)'); }
+  } else {
+    mk('brain/context.md', '# context.md — the standing facts every skill should know\n\n## Company / product\n- **Company:** <one line>\n- **Product:** <one line>\n\n## Audience & voice\n- **Primary readers:** <who>\n- **Tone:** <crisp/direct/…>\n\n## Metrics that matter\n- <the 3 numbers leadership actually tracks>\n');
+    for (const s of ['decisions', 'knowledge']) mkdir('brain/' + s);
+    for (const f of ['hypotheses', 'stakeholders', 'entities']) mk('brain/' + f + '.md', `# ${f}\n\n<!-- append dated, provenance-tagged lines: - [data] 2026-07-06 … -->\n`);
+  }
+  mkdir('brain/predictions');
+
+  // ── Standing context for the skills (pm-context.md, the older single-file form)
+  const ctxTpl = join(PKG_ROOT, 'templates', 'pm-context.example.md');
+  mk('pm-context.md', existsSync(ctxTpl) ? readFileSync(ctxTpl, 'utf8') : '# pm-context.md\n\nFill in who you are, your product, audience, and voice.\n');
+
+  // ── Arena folders the web workspace bridge writes into ─────────────────────
+  mkdir('firm-minutes');
+  mkdir('boardroom');
+
+  // ── AGENTS.md (--agents-md, or always create if missing alongside CLAUDE.md) ──
+  if (argv.includes('--agents-md') || !existsSync(join(cwd, 'AGENTS.md'))) mk('AGENTS.md', AGENTS_MD);
+
+  // ── CLAUDE.md wiring (append a section; never rewrite an existing file) ────
+  const claudeMd = join(cwd, 'CLAUDE.md');
+  if (!existsSync(claudeMd)) { mk('CLAUDE.md', '# Project notes for Claude\n' + CLAUDE_MD_SECTION); }
+  else if (readFileSync(claudeMd, 'utf8').includes('## PM Skills — this project')) kept.push('CLAUDE.md (section already present)');
+  else { if (!dryRun) appendFileSync(claudeMd, '\n' + CLAUDE_MD_SECTION); made.push('CLAUDE.md  (PM Skills section appended)'); }
+
+  // ── Report ──────────────────────────────────────────────────────────────────
+  for (const m of made) console.log(`  ✔ ${dryRun ? 'would create' : 'created'}  ${m}`);
+  for (const k of kept) console.log(`  ↷ kept       ${k}`);
+
+  console.log(`\nNext steps:
+  1. Fill in brain/context.md (2 minutes) — every skill output gets sharper.
+  2. Install the skills:        npx pm-claude-skills add --agent claude
+  3. Wire the ambient hooks:    see hooks/settings.example.json in the repo (session briefs, skill suggestions, doc quality gate).
+  4. Point the web arenas here: open the Firm or Boardroom → 🗂 Connect a folder → choose THIS directory.
+  5. Checkup anytime:           npx pm-claude-skills doctor
+
+${STAR}`);
+  return 0;
+}