A
arm-pz/skills-site
Static site and publish pipeline for an agent skills catalog: a generated landing page per SKILL.md skill or…
View on GitHub ↗Link checked 2026-11-12 · HTTP 200 · opens the vendor's own site
Install & config
Client config (built from the install command this repo gives)
{
"mcpServers": {
"arm-pz-skills-site": {
"command": "npx",
"args": ["skills"]
}
}
}Parsed from the README we fetched on 2026-10-31; we have not executed it.
command/args come from the install command the repository states. When it is ambiguous, the project README wins.
VS Code (.vscode/mcp.json - the key is servers)
{
"servers": {
"arm-pz-skills-site": {
"command": "npx",
"args": ["skills"]
}
}
}Same server, but VS Code keeps it under <code>servers</code>.
Claude Code (command line)
claude mcp add --transport stdio arm-pz-skills-site -- npx skills
Claude Code does not read a config file; add the server with this command.
Key names differ by client: Clients & CLI
- Star count
- 0 ★ · arm-pz/skills-site · read from the GitHub search API · (checked 2026-11-13 20:00:50)
- Source
- read directly from its GitHub page · type: open-source project
- Harvest history
- first seen 2026-11-01 · last seen 2026-11-14 · seen 14× · listed by 2 sources
- Score
- score 203 · 2 sources · repo
- Install
npx skills
FAQ (answers come from our records)
How do I install it?
The repository provides this install command:
npx skills. It comes from the repository itself; we cannot guarantee it runs as-is in your client.Who is it for?
We filed it under these tags: Agent 技能. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository
arm-pz/skills-site；first catalogued 2026-11-01；last seen 2026-11-14。There is no official certification or quality rating here — we record sources and dates only.# skills-site
以下摘录自该仓库的 README（片段） · 在 GitHub 读完整 README →
A static site and publish pipeline for a catalog of agent skills. Each skill
(a SKILL.md repo) or Claude Code mod gets its own landing page with a one-line
install command; /skills/, /mods/ and /workflows/ list them by type (every workflow
also gets its own page); the homepage leads with outcomes; and a
/.well-known/agent-skills/index.json discovery index lets anyone run
npx skills add https://your-domain. One GitHub Actions workflow keeps the
manifest in step with GitHub and tests, builds and verifies the site, then
commits the build to a deploy branch that Cloudflare Pages pulls through its
own GitHub app, so CI holds no Cloudflare credentials. Plain HTML, CSS and Python, no
framework. It is the source of skillsmith.tech, formerly skills.caffeinity.com.
Fork it for your own skills: list them in site/skills.json, then follow
Run locally. Every page is self-contained, and the shared design
tokens live in one :root block per page.
Related: caffeinity-skills is
the plugin marketplace for the same skills, and
publish-pipeline is the agent
skill that sets up and repairs this kind of pipeline.
## Run locally
python tools/agent-skills/build.py # needs the GitHub CLI, gh auth login
python tools/agent-skills/gensitemap.py --write
bash tools/agent-skills/makedist.sh
python -m http.server 8080 --directory dist
## Deploy
CI holds no Cloudflare credentials; Cloudflare pulls. Every hour, and on every
push to main, .github/workflows/publish.yml runs sync.py, which brings
site/skills.json in line with GitHub (new skill repos, moved repo HEADs,
renames, removals), rebuilds the site, gates it with checkdist.py and
verify.mjs, commits the sync back to main, and commits the gated dist/ to
the deploy branch (under public/, only when the build changed). The
Cloudflare Pages project is connected to that branch, so each push there goes
live in a few minutes. The job then waits for the live version.json to match
and verifies the live site; if it never matches it opens one "Deploy pending"
issue, and closes it once live matches. .github/workflows/health.yml runs tune-skill's regression check over
every published skill each night. PROJECT.md maps the whole ecosystem.
### Connecting Cloudflare Pages (once)
Cloudflare cannot attach Git to a project created by direct upload, so this is
a new Pages project that takes over the domain:
1. Merge to main and let publish run once, so the deploy branch exists.
2. Cloudflare dashboard → Workers & Pages → Create → Pages → Connect to Git.
Authorize the Cloudflare GitHub app for arm-pz/skills-site only.
3. Production branch deploy; framework preset None; build command empty
(or exit 0); build output directory public. No environment variables.
4. Settings → Builds → Branch control: Preview branches: None, so main
and PR branches never build (they hold sources, not a site).
5. Once its .pages.dev URL serves the hub, remove skills.caffeinity.com
from the old skills project's Custom domains and add it to the new one.
6. Run publish again (Actions → publish → Run workflow); the "Deploy pending"
(a SKILL.md repo) or Claude Code mod gets its own landing page with a one-line
install command; /skills/, /mods/ and /workflows/ list them by type (every workflow
also gets its own page); the homepage leads with outcomes; and a
/.well-known/agent-skills/index.json discovery index lets anyone run
npx skills add https://your-domain. One GitHub Actions workflow keeps the
manifest in step with GitHub and tests, builds and verifies the site, then
commits the build to a deploy branch that Cloudflare Pages pulls through its
own GitHub app, so CI holds no Cloudflare credentials. Plain HTML, CSS and Python, no
framework. It is the source of skillsmith.tech, formerly skills.caffeinity.com.
Fork it for your own skills: list them in site/skills.json, then follow
Run locally. Every page is self-contained, and the shared design
tokens live in one :root block per page.
Related: caffeinity-skills is
the plugin marketplace for the same skills, and
publish-pipeline is the agent
skill that sets up and repairs this kind of pipeline.
## Run locally
python tools/agent-skills/build.py # needs the GitHub CLI, gh auth login
python tools/agent-skills/gensitemap.py --write
bash tools/agent-skills/makedist.sh
python -m http.server 8080 --directory dist
## Deploy
CI holds no Cloudflare credentials; Cloudflare pulls. Every hour, and on every
push to main, .github/workflows/publish.yml runs sync.py, which brings
site/skills.json in line with GitHub (new skill repos, moved repo HEADs,
renames, removals), rebuilds the site, gates it with checkdist.py and
verify.mjs, commits the sync back to main, and commits the gated dist/ to
the deploy branch (under public/, only when the build changed). The
Cloudflare Pages project is connected to that branch, so each push there goes
live in a few minutes. The job then waits for the live version.json to match
and verifies the live site; if it never matches it opens one "Deploy pending"
issue, and closes it once live matches. .github/workflows/health.yml runs tune-skill's regression check over
every published skill each night. PROJECT.md maps the whole ecosystem.
### Connecting Cloudflare Pages (once)
Cloudflare cannot attach Git to a project created by direct upload, so this is
a new Pages project that takes over the domain:
1. Merge to main and let publish run once, so the deploy branch exists.
2. Cloudflare dashboard → Workers & Pages → Create → Pages → Connect to Git.
Authorize the Cloudflare GitHub app for arm-pz/skills-site only.
3. Production branch deploy; framework preset None; build command empty
(or exit 0); build output directory public. No environment variables.
4. Settings → Builds → Branch control: Preview branches: None, so main
and PR branches never build (they hold sources, not a site).
5. Once its .pages.dev URL serves the hub, remove skills.caffeinity.com
from the old skills project's Custom domains and add it to the new one.
6. Run publish again (Actions → publish → Run workflow); the "Deploy pending"
Listed badge (optional)
Copy this line into the README if you want the listing mark. Optional, and it does not affect being listed.
[](https://anclum.com/en/s/arm-pz-skills-site/)
- Source
- github · https://github.com/arm-pz/skills-site
- Browse by field
- Agent skills
- Listed by
- 2 independent sources
- added
- today
More like this
Filed under
Every tag opens the full list for that filter. Counts are the live catalogue figures.