Skip to content
Back to search

skillnote

The open-source skill registry for AI coding agents. Create, manage, and distribute SKILL.md files across Openclaw, Claude Code, Cursor, Codex, OpenHands, Antigravity, and more.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked yesterday
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/luna-prompts--skillnote/badge.svg)](/listings/luna-prompts--skillnote)

HTML

<a href="/listings/luna-prompts--skillnote"><img src="/api/listings/luna-prompts--skillnote/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

npx skillnote start

Opens <http://localhost:3000>.

Requires **Docker** (running) and **Node.js 20+**. The CLI pulls the published images from GHCR, brings up the web + API + Postgres stack, waits for healthchecks, and opens the dashboard. About 30 seconds on a warm cache.

┌────────────────────────────────────────────┐
│  SkillNote ▸  the skill registry for AI    │
│  v0.5.2 · github.com/luna-prompts/skillnote│
└────────────────────────────────────────────┘
◇  Prerequisites ok
◇  Images pulled
◇  Containers running
◇  Services healthy

╭────────┬──────────────────────────────╮
│ Web UI │ http://localhost:3000        │
│ API    │ http://localhost:8082        │
╰────────┴──────────────────────────────╯

<details> <summary><b>Other ways to install</b></summary>

**Docker Compose directly** (no Node):

curl -fsSL https://raw.githubusercontent.com/luna-prompts/skillnote/cli-v0.5.4/deploy/docker-compose.yml -o docker-compose.yml
docker compose up -d

**LAN-accessible** for teammates on the same network:

SKILLNOTE_HOST=<your-lan-ip> docker compose up -d

**Podman**: `npx skillnote start` is Docker-only, but the Compose path and `install.sh` both work with `podman compose` (Podman 4+) or `podman-compose`.

**Building from source** (contributors):

git clone https://github.com/luna-prompts/skillnote.git
cd skillnote && ./install.sh

`install.sh` builds from local source instead of pulling published images. Auto-detects `docker compose`, `podman compose`, or `podman-compose`.

</details>

Lifecycle commands

npx skillnote start         # boot + open UI
npx skillnote stop          # halt; volumes preserved
npx skillnote restart       # stop + start
npx skillnote status        # health table (--json for scripts)
npx skillnote logs [svc]    # tail logs (-f to follow)
npx skillnote open          # open UI (--app for chromeless)
npx skillnote doctor        # 11 health checks
npx skillnote reset --confirm   # DESTRUCTIVE: drops all data

---

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 525 lines · 2 findings marked

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <p align="center">
  2. <img src="public/logo.svg" width="88" height="88" alt="SkillNote" />
  3. </p>
  4. <h1 align="center">S K I L L N O T E</h1>
  5. <p align="center">
  6. <strong>The open-source skill registry for AI coding agents.</strong>
  7. <br />
  8. Self-host your team's <code>SKILL.md</code> library. Version it, scope it, and ship it to Claude Code and OpenClaw from one CLI.
  9. </p>
  10. <p align="center">
  11. <a href="https://www.npmjs.com/package/skillnote"><img src="https://img.shields.io/npm/v/skillnote.svg?label=npm&color=cb3837&logo=npm" alt="npm version" /></a>
  12. <a href="https://www.npmjs.com/package/skillnote"><img src="https://img.shields.io/npm/dt/skillnote.svg?label=downloads&color=brightgreen" alt="npm total downloads" /></a>
  13. <a href="https://github.com/luna-prompts/skillnote/pkgs/container/skillnote-api"><img src="https://img.shields.io/badge/ghcr.io-multi--arch-2496ED?logo=docker&logoColor=white" alt="Docker images on GHCR" /></a>
  14. <a href="https://clawhub.ai/latentloop07/skillnote"><img src="https://img.shields.io/badge/clawhub-skillnote-F97316" alt="On clawhub" /></a>
  15. <a href="https://github.com/luna-prompts/skillnote/blob/master/LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue.svg" alt="License" /></a>
  16. <a href="https://discord.gg/GazU4amU6H"><img src="https://img.shields.io/badge/Discord-Join%20us-5865F2?logo=discord&logoColor=white" alt="Discord" /></a>
  17. <a href="https://github.com/luna-prompts/skillnote"><img src="https://img.shields.io/github/stars/luna-prompts/skillnote?style=social" alt="Stars" /></a>
  18. </p>
  19. <p align="center">
  20. <a href="#quick-start"><strong>Quick start</strong></a> &nbsp;·&nbsp;
  21. <a href="#the-8000-character-problem">Why</a> &nbsp;·&nbsp;
  22. <a href="#features">Features</a> &nbsp;·&nbsp;
  23. <a href="#agent-support">Agents</a> &nbsp;·&nbsp;
  24. <a href="#architecture">Architecture</a> &nbsp;·&nbsp;
  25. <a href="#faq">FAQ</a> &nbsp;·&nbsp;
  26. <a href="https://discord.gg/GazU4amU6H">Discord</a>
  27. </p>
  28. <br />
  29. <p align="center">
  30. <img src="docs/screenshots/connect-browse.png" width="100%" alt="SkillNote Connect page browsing Claude Code, OpenClaw, and claude.ai as official agent integrations, each shown with its canonical mark" />
  31. </p>
  32. ---
  33. ## The 8,000-character problem
  34. Claude Code shares **~8,000 characters** across every active skill description ([docs](https://docs.anthropic.com/en/docs/claude-code/skills)). Past that limit, descriptions silently truncate. The system prompt forbids using skills that aren't listed in context, so truncated skills are both invisible *and* explicitly off-limits ([#13343](https://github.com/anthropics/claude-code/issues/13343), [#40121](https://github.com/anthropics/claude-code/issues/40121)).
  35. In practice, past ~15 active skills your skills stop working and you can't tell which ones. New teammates have no way to discover what skills the project depends on. Updating a shared skill means re-zipping and re-uploading for everyone. And private skills, like deploy procedures, compliance workflows, or internal API patterns, have nowhere safe to live.
  36. **SkillNote is a self-hosted registry that fixes that.** Per-project collections scope which skills load. Live sync pushes browser edits to every connected agent within 60 seconds. Agents rate skills 1 to 5 after using them, so you finally have signal on what works. Your skills stay on your infrastructure. Your servers, your rules.
  37. | Without SkillNote | With SkillNote |
  38. | --- | --- |
  39. | Skills truncate past ~15 active | Collections scope to 15 per project |
  40. | `~/.claude/skills/` per laptop | One registry, every agent |
  41. | Re-zip + re-upload to share an edit | Edit in browser, every session picks it up in 60s |
  42. | No signal on what actually works | Agents rate every skill they use, 1-5 + comment |
  43. | Private skills have nowhere safe | Self-hosted; never leaves your network |
  44. ---
  45. ## Quick start
  46. ```bash
  47. npx skillnote start
  48. ```
  49. Opens <http://localhost:3000>.
  50. Requires **Docker** (running) and **Node.js 20+**. The CLI pulls the published images from GHCR, brings up the web + API + Postgres stack, waits for healthchecks, and opens the dashboard. About 30 seconds on a warm cache.
  51. ```text
  52. ┌────────────────────────────────────────────┐
  53. │ SkillNote ▸ the skill registry for AI │
  54. │ v0.5.2 · github.com/luna-prompts/skillnote│
  55. └────────────────────────────────────────────┘
  56. ◇ Prerequisites ok
  57. ◇ Images pulled
  58. ◇ Containers running
  59. ◇ Services healthy
  60. ╭────────┬──────────────────────────────╮
  61. │ Web UI │ http://localhost:3000 │
  62. │ API │ http://localhost:8082 │
  63. ╰────────┴──────────────────────────────╯
  64. ```
  65. <details>
  66. <summary><b>Other ways to install</b></summary>
  67. **Docker Compose directly** (no Node):
  68. ```bash
  69. curl -fsSL https://raw.githubusercontent.com/luna-prompts/skillnote/cli-v0.5.4/deploy/docker-compose.yml -o docker-compose.yml
  70. docker compose up -d
  71. ```
  72. **LAN-accessible** for teammates on the same network:
  73. ```bash
  74. SKILLNOTE_HOST=<your-lan-ip> docker compose up -d
  75. ```
  76. **Podman**: `npx skillnote start` is Docker-only, but the Compose path and `install.sh` both work with `podman compose` (Podman 4+) or `podman-compose`.
  77. **Building from source** (contributors):
  78. ```bash
  79. git clone https://github.com/luna-prompts/skillnote.git
  80. cd skillnote && ./install.sh
  81. ```
  82. `install.sh` builds from local source instead of pulling published images. Auto-detects `docker compose`, `podman compose`, or `podman-compose`.
  83. </details>
  84. ### Lifecycle commands
  85. ```text
  86. npx skillnote start # boot + open UI
  87. npx skillnote stop # halt; volumes preserved
  88. npx skillnote restart # stop + start
  89. npx skillnote status # health table (--json for scripts)
  90. npx skillnote logs [svc] # tail logs (-f to follow)
  91. npx skillnote open # open UI (--app for chromeless)
  92. npx skillnote doctor # 11 health checks
  93. npx skillnote reset --confirm # DESTRUCTIVE: drops all data
  94. ```
  95. ---
  96. ## Wire up your AI agent
  97. After the backend is running, install the plugin for your agent.
  98. ### Claude Code
  99. ```bash
  100. npx skillnote connect claude-code
  101. source ~/.zshrc # or ~/.bashrc
  102. ```
  103. Runs the canonical `/setup/agent` script: registers the plugin marketplace in `~/.claude/settings.json`, installs the SkillNote plugin into `~/.claude/plugins/`, drops picker binaries in `~/.skillnote/bin/`, and adds a shell wrapper. Run `claude` in any project, and the collection picker appears on first launch:
  104. <p align="center">
  105. <img src="docs/terminal/picker4.png" width="640" alt="SkillNote collection picker in Claude Code terminal showing Conventions, DevOps, and Frontend collections" />
  106. </p>
  107. Pick a collection (it's saved to `.skillnote.json`) and your scoped skills load on every session.
  108. <details>
  109. <summary><i>Or, paste this prompt into a fresh Claude Code session</i></summary>
  110. <br />
  111. ```text
  112. Install SkillNote on my machine and wire it into this Claude Code session.
  113. 1. If http://localhost:8082/health doesn't respond, run:
  114. npx skillnote start --no-browser -d
  115. 2. If ~/.claude/plugins/skillnote/ doesn't exist, run:
  116. curl -sf http://localhost:8082/setup | bash
    • Note: Pipes a downloaded script directly into a shell (curl | sh / wget | bash)
    • Line 159 on GitHub
  117. 3. source ~/.zshrc
  118. 4. Confirm: claude --version, ls ~/.claude/plugins/skillnote/
  119. 5. Report what collection picker options you see when running `claude`.
  120. ```
  121. </details>
  122. ### OpenClaw
  123. ```bash
  124. clawhub install skillnote
  125. ```
  126. That's the whole install for the default `localhost:8082` setup. If the backend isn't running, the skill auto-bootstraps it on first sync.
  127. The skill ships `sync.sh` (60s catalog sync), `log-watcher.py` (analytics daemon), `install-backend.sh` (bootstrap), and an always-loaded `SKILL.md` that grafts a persistent `<skillnote v1>` block into `~/.openclaw/workspace/AGENTS.md`.
  128. <details>
  129. <summary><i>Non-default host or alternative install</i></summary>
  130. <br />
  131. For a non-default host:
  132. ```bash
  133. export SKILLNOTE_BASE_URL="http://your-server:8082"
  134. clawhub install skillnote
  135. echo 'export SKILLNOTE_BASE_URL="http://your-server:8082"' >> ~/.zshrc
    • Dismissed after review: Sets up something that runs again on its own (cron, launchd, systemd, startup entry or shell profile) and fetches from the network
    • Line 187 on GitHub
  136. ```
  137. Or scripted (no clawhub):
  138. ```bash
  139. npx skillnote start --no-browser -d
  140. npx skillnote connect openclaw
  141. ```
  142. </details>
  143. ### claude.ai (web app)
  144. A small Chrome extension syncs your SkillNote collections to your claude.ai account as plugin groups — skills you publish in SkillNote appear in claude.ai's **Customize → Plugins**, and claude.ai-authored skills import back. No copy-paste either way. The extension opens as a **side panel** beside claude.ai, matches its light/dark theme, and shows how often your skills actually get used.
  145. <div align="center">
  146. <img src="docs/screenshots/claude-ai-connector.png" width="360" alt="SkillNote claude.ai connector side panel: Connected status, a 'This week on claude.ai' usage card (42 skill uses, most used code-review-checklist), and the collections currently live on claude.ai" />
  147. </div>
  148. **Install** (Chrome Web Store listing pending review — load unpacked for now):
  149. ```bash
  150. cd extensions/claude-ai && npm install && npm run build
  151. # chrome://extensions → enable Developer mode → Load unpacked → select dist/
  152. ```
  153. **Connect** — everything happens in the panel, no separate tab:
  154. 1. Click the SkillNote toolbar icon to open the side panel.
  155. 2. Enter your SkillNote URL — the same address you open in your browser (e.g. `http://localhost:3000`) — and click **Connect**. (Chrome asks once for permission to reach that address.)
  156. 3. In SkillNote, the **notifications bell** (top-right) shows the pairing request — confirm the code matches and click **Approve**.
  157. 4. Choose what to sync: on any collection in SkillNote, open **Sync ▾ → claude.ai** and flip the toggle. Those skills appear in claude.ai within seconds (as the plugin group `SkillNote: <collection>`), and re-sync automatically on every change.
  158. <div align="center">
  159. <img src="docs/screenshots/collection-sync-claude-ai.png" width="100%" alt="A SkillNote collection's Sync menu open, showing the claude.ai connector toggled on — the collection is live on claude.ai as the plugin group 'SkillNote: conventions', with OpenAI listed as coming soon" />
  160. </div>
  161. You stay in control of what leaves SkillNote: the **Sync** menu is per-collection, so dev-only or sensitive collections simply stay off. Toggle one on and it's live on claude.ai; toggle it off and the connector retires that plugin group.
  162. Sync runs automatically while you're signed in to claude.ai. The extension reads your claude.ai session cookies **locally only** — they never leave your machine, and it only ever talks to claude.ai and the SkillNote URL you entered.
  163. Full walkthrough: [`docs/claude-ai-user-guide.md`](docs/claude-ai-user-guide.md) · architecture: [`docs/claude-ai-integration.md`](docs/claude-ai-integration.md) · admin runbook: [`docs/claude-ai-admin-runbook.md`](docs/claude-ai-admin-runbook.md) · privacy: [`extensions/claude-ai/PRIVACY.md`](extensions/claude-ai/PRIVACY.md).
  164. > Cursor, Codex, Antigravity, and OpenHands are on the roadmap. [Open an issue](https://github.com/luna-prompts/skillnote/issues) if you want to help build an adapter.
  165. ---
  166. ## Features
  167. ### Per-project collections
  168. Claude Code shares ~8,000 characters across every active skill description; past ~15 skills, descriptions silently truncate and truncated skills won't trigger. Collections scope which skills load per project: frontend project gets React + testing patterns, API project gets error handling + deploy conventions. Same registry, different active sets, no context wasted.
  169. <p align="center">
  170. <img src="docs/screenshots/collections.png" width="100%" alt="Collections page with skill counts and progress bars" />
  171. </p>
  172. If your folder name matches a collection, the plugin recommends it automatically.
  173. ### Import from any GitHub repo
  174. The community has published thousands of `SKILL.md` files since Anthropic released the format. Paste a GitHub URL, shorthand (`garrytan/gstack`), a tree URL to a subfolder, or a Claude Code marketplace manifest (`anthropic.json`). SkillNote shallow-clones the repo, scans every `SKILL.md`, validates frontmatter, and opens a workspace where you pick exactly what to install.
  175. <p align="center">
  176. <img src="docs/screenshots/marketplace-workspace.png" width="100%" alt="Marketplace workspace showing skills from garrytan/gstack with select-all, filter, and per-skill preview pane" />
  177. </p>
  178. Some popular registries to try:
  179. - [**`anthropics/skills`**](https://github.com/anthropics/skills): Anthropic's official Agent Skills repository
  180. - [**`ComposioHQ/awesome-claude-skills`**](https://github.com/ComposioHQ/awesome-claude-skills): 800+ community skills, the largest curated set
  181. - [**`alirezarezvani/claude-skills`**](https://github.com/alirezarezvani/claude-skills): 600+ skills for Claude Code, Codex, Gemini CLI, Cursor, and more
  182. - [**`garrytan/gstack`**](https://github.com/garrytan/gstack): Garry Tan's 50+ opinionated YC-flavored tools (CEO, Designer, Eng Manager, etc.)
  183. - [**`obra/superpowers`**](https://github.com/obra/superpowers): Jesse Vincent's agentic skills framework
  184. ### Live sync, every agent
  185. Edit a skill in the browser and every running Claude Code or OpenClaw session picks up the change within 60 seconds. Claude Code re-syncs on every prompt and hot-reloads `SKILL.md` mid-session. OpenClaw's `sync.sh` runs on a 60s throttle. One person updates a skill, everyone gets it. New teammates run the setup command once and inherit every skill the team has built.
  186. ### Agent reviews
  187. Most skill setups are fire-and-forget. SkillNote closes the loop. After applying a skill, the agent rates it 1-5 and describes what it did. OpenClaw additionally posts a `linked_usage_id` correlating each rating to the specific task that produced it. You see which skills are actually being used, which ones break, and how performance changes across versions.
  188. <p align="center">
  189. <img src="docs/screenshots/analytics-dashboard.png" width="100%" alt="Analytics dashboard with total calls, unique skills, agent breakdown, and skill leaderboard" />
  190. </p>
  191. ### Version history
  192. Every save creates a snapshot. Browse, compare, and restore any version in one click. Published versions use semver and ship as checksummed ZIP bundles.
  193. ### Skill push
  194. When Claude Code notices you correct the same thing three times ("use pnpm, not npm"), it offers to turn it into a skill. The skill is pushed to SkillNote and syncs to every connected agent in 60 seconds. What one person teaches once becomes a skill everyone has.
  195. ---
  196. ## Agent support
  197. | Agent | Status | Mechanism |
  198. | --- | --- | --- |
  199. | **Claude Code** | Supported | Native plugin (`~/.claude/plugins/skillnote/`) with 6 lifecycle hooks |
  200. | **OpenClaw** | Supported | clawhub skill bundle with `sync.sh` + analytics daemon |
  201. | Cursor | Planned | Roadmap |
  202. | Codex CLI | Planned | Roadmap |
  203. | Antigravity | Planned | Roadmap |
  204. | OpenHands | Planned | Roadmap |
  205. Want to help build an adapter? [Open an issue](https://github.com/luna-prompts/skillnote/issues) or join us on [Discord](https://discord.gg/GazU4amU6H).
  206. ---
  207. ## Architecture
  208. ```
  209. ┌──────────────────────────────────────────────────────┐
  210. │ │
  211. │ SkillNote Server (Docker) │
  212. │ │
  213. │ Web UI REST API PostgreSQL │
  214. │ :3000 :8082 (storage + notify) │
  215. │ │
  216. └────────────────────┬─────────────────────────────────┘
  217. │
  218. REST API
  219. │
  220. ┌────────────┴────────────┐
  221. ▼ ▼
  222. ┌────────────────────┐ ┌────────────────────┐
  223. │ Claude Code plugin │ │ OpenClaw skill │
  224. │ │ │ │
  225. │ ~/.claude/ │ │ ~/.openclaw/ │
  226. │ plugins/ │ │ skills/ │
  227. │ skillnote/ │ │ skillnote/ │
  228. │ │ │ │
  229. │ 6 lifecycle hooks │ │ sync.sh + daemon │
  230. │ Per-project picker │ │ AGENTS.md graft │
  231. └────────────────────┘ └────────────────────┘
  232. ```
  233. SkillNote uses each agent's native skill system. For Claude Code that's hooks plus plugin format, with `SessionStart`, `UserPromptSubmit`, `PostToolUse`, `PostCompact`, `SubagentStart`, and `Stop`. Only `SessionStart` blocks (for ~1 second to sync); every other hook runs async, so you never wait for SkillNote.
  234. For OpenClaw it's a clawhub-installable bundle with `sync.sh` (catalog), `log-watcher.py` (analytics), and an AGENTS.md graft that keeps the agent consulting the registry on every task.
  235. Skills are written as local `SKILL.md` files, not piped through a network abstraction. That means every [Claude Code frontmatter feature](https://docs.anthropic.com/en/docs/claude-code/skills), including `allowed-tools`, `context: fork`, `effort`, and `model`, works natively. These features only work with on-disk `SKILL.md` files, which is why SkillNote syncs to disk instead of serving skills over a network protocol.
  236. For the full HLD see [`docs/openclaw-hld.md`](docs/openclaw-hld.md).
  237. ---
  238. ## SKILL.md format
  239. ```markdown
  240. ---
  241. name: pdf-extractor
  242. description: Extract text and tables from PDF files. Use when the user mentions PDFs or scanned documents.
  243. collections: [data, documents]
  244. allowed-tools: Read Write Bash(pdftotext *)
  245. context: fork
  246. ---
  247. # PDF Extractor
  248. When the user provides a PDF file:
  249. 1. Use `pdftotext` to extract raw text
  250. 2. Identify tables and format them as markdown
  251. 3. Preserve headings and document structure
  252. ```
  253. ---
  254. ## Security & deployment
  255. SkillNote is built for trusted environments: a developer's laptop, a team VM on a private network, or a self-hosted server behind a VPN. Out of the box it has **no authentication** on the web UI or API; anything that can reach `:3000` and `:8082` can read and write skills.
  256. - **Local-only (default):** `npx skillnote start` binds to `localhost`. Safe.
  257. - **LAN-only:** set `SKILLNOTE_HOST=<lan-ip>` to expose to teammates on the same network. Assumes the LAN is trusted.
  258. - **Internet-exposed:** never bind `:3000` or `:8082` directly to a public IP. Put it behind a reverse proxy (Caddy, Nginx, Traefik) with basic auth, OAuth, or a Tailscale/Cloudflare Tunnel.
  259. - **Marketplace imports:** every install writes `SKILL.md` files that your agent will read. Review the workspace preview before importing from unfamiliar sources.
  260. Auth on the API is on the roadmap. Until then, treat reachability as the access boundary.
  261. ---
  262. ## Tech stack
  263. | Layer | Technology |
  264. | --- | --- |
  265. | CLI | Node 20+, TypeScript, commander, [@clack/prompts](https://www.npmjs.com/package/@clack/prompts) |
  266. | Frontend | Next.js 16, React 19, TypeScript, Tailwind CSS 4, Tiptap, PWA |
  267. | Backend | Python 3.12, FastAPI, SQLAlchemy 2, Alembic |
  268. | Claude Code plugin | Bash, Python, Claude Code Plugin API |
  269. | OpenClaw skill | Bash (`sync.sh`), Python (`log-watcher.py`), clawhub bundle |
  270. | Database | PostgreSQL 16 |
  271. | Distribution | npm (`skillnote`), GHCR multi-arch images, Docker Compose |
  272. ---
  273. ## FAQ
  274. <details>
  275. <summary><b>What is SkillNote?</b></summary>
  276. <br />
  277. SkillNote is an open-source, self-hosted registry for Anthropic's [Agent Skills format](https://docs.anthropic.com/en/docs/claude-code/skills). You run it on your own infrastructure (a laptop, a team VM, a private server), and it gives your team one place to create, version, and share `SKILL.md` files across Claude Code, OpenClaw, and other coding agents. The web UI lets you edit skills in a Notion-style editor; the CLI plugin syncs your registry into each agent's native skill system so the skills hot-reload mid-session.
  278. </details>
  279. <details>
  280. <summary><b>How is SkillNote different from MCP?</b></summary>
  281. <br />
  282. MCP (Model Context Protocol) is a wire protocol for talking to tools over a network. Skills are local `SKILL.md` files that the agent reads from disk. The two solve different problems: MCP lets an agent call a remote service; skills tell an agent *how* to do something with instructions, examples, and frontmatter directives like `allowed-tools` or `context: fork`. SkillNote distributes the second kind. It syncs skills to disk because that's where features like `allowed-tools`, `context: fork`, `effort`, and `model` are actually evaluated by Claude Code, and they don't work over a network protocol.
  283. </details>
  284. <details>
  285. <summary><b>How do I share Claude Code skills across my team?</b></summary>
  286. <br />
  287. Run SkillNote on a server your team can reach (`SKILLNOTE_HOST=<lan-ip> docker compose up -d` for a LAN, or behind a reverse proxy for a remote team). Everyone installs the SkillNote plugin into Claude Code with `npx skillnote connect claude-code`. From then on, any edit anyone makes in the SkillNote web UI propagates to every running Claude Code session in 60 seconds. New teammates run the connect command once and inherit every skill the team has built.
  288. </details>
  289. <details>
  290. <summary><b>Is SkillNote free?</b></summary>
  291. <br />
  292. Yes. SkillNote is MIT licensed and self-hosted. There is no cloud tier, no paid plan, and no usage telemetry. You run it on your own machines and own your data.
  293. </details>
  294. <details>
  295. <summary><b>Claude Code: skills from another collection are showing up</b></summary>
  296. <br />
  297. Claude Code discovers skills from parent directories. If you ran `claude` in a parent folder (`~/projects/`) and picked a collection, those skills persist in `~/projects/.claude/skills/` and leak into every subdirectory project.
  298. ```bash
  299. rm -rf ~/path/to/parent/.claude/skills/skillnote-*
  300. ```
  301. Always run `claude` from the actual project directory, not from umbrella folders.
  302. </details>
  303. <details>
  304. <summary><b>Claude Code: plugin changes not taking effect</b></summary>
  305. <br />
  306. Claude Code loads plugins at startup. If you reinstall the plugin while Claude Code is running, quit and restart it for the new plugin to load.
  307. </details>
  308. <details>
  309. <summary><b>OpenClaw: no <code>sn-*</code> skills appear after install</b></summary>
  310. <br />
  311. Check that `sync.sh` ran successfully on first session:
  312. ```bash
  313. ~/.openclaw/skills/skillnote/sync.sh
  314. ls ~/.openclaw/skills/sn-* | head
  315. ```
  316. If sync produces nothing, verify `~/.openclaw/skills/skillnote/config.json` has a `host` field pointing at a reachable backend (`curl -sf $host/health`).
  317. </details>
  318. <details>
  319. <summary><b>OpenClaw: agent stopped using SkillNote mid-conversation</b></summary>
  320. <br />
  321. The `<skillnote v1>` block in `~/.openclaw/workspace/AGENTS.md` is what tells the agent to consult the registry on every task. If it gets removed (manual edit, regenerated by another tool), `sync.sh` re-grafts it on the next run.
  322. ```bash
  323. grep -c '<skillnote v1>' ~/.openclaw/workspace/AGENTS.md
  324. # expected: 1
  325. ```
  326. If `0`, run `~/.openclaw/skills/skillnote/sync.sh` once and re-check.
  327. </details>
  328. <details>
  329. <summary><b>OpenClaw: analytics daemon isn't running</b></summary>
  330. <br />
  331. ```bash
  332. PID=$(cat ~/.openclaw/skills/skillnote/.log-watcher.pid 2>/dev/null) && kill -0 $PID 2>/dev/null && echo alive || echo dead
  333. ```
  334. If dead, the next `sync.sh` run relaunches it.
  335. </details>
  336. ---
  337. ## Contributing
  338. 1. Fork the repo
  339. 2. `git checkout -b feat/my-feature`
  340. 3. Commit with [Conventional Commits](https://www.conventionalcommits.org/)
  341. 4. Push and open a PR
  342. [`CLAUDE.md`](CLAUDE.md) has the architectural patterns and conventions. [`docs/`](docs/) has design docs and HLDs.
  343. Active development happens on [Discord](https://discord.gg/GazU4amU6H).
  344. ---
  345. ## Contributors
  346. <a href="https://github.com/luna-prompts/skillnote/graphs/contributors">
  347. <img src="https://contrib.rocks/image?repo=luna-prompts/skillnote" alt="Contributors" />
  348. </a>
  349. ---
  350. ## License
  351. MIT &copy; [Luna Prompts](https://github.com/luna-prompts)
  352. <p align="center">
  353. <br />
  354. <a href="https://github.com/luna-prompts/skillnote"><img src="https://img.shields.io/github/stars/luna-prompts/skillnote?style=for-the-badge&logo=github&label=Star%20us" alt="Star us" /></a>
  355. <br /><br />
  356. Built by <a href="https://github.com/luna-prompts"><strong>Luna Prompts</strong></a>
  357. </p>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.