agentleFS
SonghongleiSonghonglei/build-better-skills/skills/skill-introduction/SKILL.md

skill-introduction

Skill3 starsChanged 7 days agoMITView on GitHub

Generate a beautiful, deployable HTML introduction page for any AgentSkill. Reads USAGE.md (preferred) or SKILL.md, parses name, description, and feature sections, then renders a polished page with a hero section, feature cards, quick start, command examples, collapsible doc cards, side navigation, and a floating install button. Supports 4 themes: light (default), aurora (dark glass), techblue (debug-tool style), finance (dark blue + gold). Optionally deploys via a pluggable hook so re-runs update the same URL. Use when the user says "generate a skill intro page", "make a doc page for this skill", "publish this skill's docs", "skill-introduction". If the user does not specify a target skill, ask them for the skill name or directory path.

  • Installs packages

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Default resolution: --source > USAGE.md > SKILL.md (fallback with notice).

    line 57
  2. Distinctive:

    After successful deploy, the dashboard id is cached so subsequent runs auto-update the same URL.

    line 125
  3. Very specific:

    If no hook is configured, the skill prints [INFO] No deploy hook configured and leaves the local HTML untouched.

    line 146
  4. Distinctive:

    If the user does not provide a skill path, ask them.

    line 153

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

skill-introduction

Auto-generate a polished HTML introduction page from a skill's USAGE.md or SKILL.md. Four themes, optional deploy hook, re-runs update the same URL.

Quick Start

Generate only (no deploy)

python3 scripts/generate_html.py --skill-dir <path-to-skill> --no-deploy

Outputs to ./output/<slug>-intro.html (override with --output).

Specify author

python3 scripts/generate_html.py --skill-dir <path> --author "Jane Doe"

Author resolution order: --author → SKILL_INTRO_AUTHOR env → git config --global user.name → $USER.

Pick a theme

python3 scripts/generate_html.py --skill-dir <path> --theme aurora

Themes: light (default), aurora (dark glass), techblue (debug-tool), finance (dark blue + gold).

Use a specific source file

python3 scripts/generate_html.py --skill-dir <path> --source USAGE.md

Default resolution: --source > USAGE.md > SKILL.md (fallback with notice).

Custom subtitle (recommended)

python3 scripts/generate_html.py --skill-dir <path> \
  --subtitle "Beautiful, deployable HTML docs for your skill"

Without --subtitle, the page truncates the long description field — usually ugly. Always pass a one-liner.

Generated Page Layout

RegionContent
HeroSkill name, subtitle, trigger tags, install/docs buttons
HighlightsAuto-extracted from ## sections, up to 6 cards (hidden if none)
FeaturesColored top-border cards, 3-col grid, up to 6 (hidden if none)
Quick StartTabs: user steps + terminal commands
Detailed DocsTabs: collapsible doc cards grouped by h2
RoadmapShown only if source has "roadmap/plan/future/todo" sections
Known IssuesExtracted from "notes/issues/limitations/known" sections
FooterSkill name · maintainer · last updated
Side navRight-fixed, scroll-spy highlight
Floating installBottom-right, links to install URL

Parameters

FlagDescriptionDefault
--skill-dirTarget skill directory (required)—
--outputOutput HTML path./output/<name>-intro.html
--no-deployGenerate only, skip deploydeploy if hook configured
--hub-urlInstall button URLhttps://clawhub.com/skill/<slug>
--authorMaintainer nameresolved from env/git/$USER
--update-idForce-update specific dashboard idfrom cache
--themelight / aurora / techblue / financelight
--sourceSource md file (relative or absolute)auto USAGE.md > SKILL.md
--subtitleHero subtitle one-linertruncates description

Source Selection (important)

Selection order:

  1. --source (explicit)
  2. USAGE.md in skill dir (recommended, user-facing)
  3. SKILL.md (fallback)

⚠️ SKILL.md is written for AI trigger matching (contains technical triggers, "when the user says…" phrasing). Rendering it directly to humans looks technical. If falling back to SKILL.md, the page shows a notice at the top and the terminal prints [WARN]. Add a USAGE.md for a polished user-facing page. Never soften SKILL.md's description to make it pretty — that breaks AI trigger matching.

Subtitle (agents: do this)

The hero subtitle should be a one-liner pitch, not raw description. Descriptions are long (contain triggers + technical detail); truncating to 130 chars + ... looks bad.

Agents: read the skill, craft a concise pitch, pass it via --subtitle.

Examples:

  • ❌ Truncated: Generate a beautiful, deployable HTML introduction page for any AgentSkill. Reads USAGE.md (preferred) or SKILL.md, parses name...
  • ✅ One-liner: Beautiful, deployable HTML docs for your skill in one command

Rendering Engine

Block-level markdown (tables / nested lists / blockquotes / hr) is rendered by vendored mistune (BSD-3-Clause, in vendor/mistune/, ships with the skill, pure Python). If vendor/ is missing the script falls back to a built-in regex renderer — no crash, just simpler output.

Cache

After successful deploy, the dashboard id is cached so subsequent runs auto-update the same URL.

SettingDefaultOverride
Cache file~/.cache/skill-introduction/cache.jsonSKILL_INTRO_CACHE env
Authorenv > git > $USERSKILL_INTRO_AUTHOR env
Deploy hooknoneSKILL_INTRO_DEPLOY_CMD env (see below)

Deploy Hook (optional)

The skill does not bind to any hosting platform. To enable deploy, set SKILL_INTRO_DEPLOY_CMD to a script/command that:

  1. Accepts the HTML file path as $1
  2. (Optional) reads SKILL_INTRO_UPDATE_ID env when updating an existing page
  3. Prints the deployed URL (containing dashboardId=<32hex>) on stdout to enable cache-based updates
export SKILL_INTRO_DEPLOY_CMD="/path/to/my-deploy.sh"
python3 scripts/generate_html.py --skill-dir <path>

If no hook is configured, the skill prints [INFO] No deploy hook configured and leaves the local HTML untouched.

Notes

  • Target skill must have USAGE.md or SKILL.md or the script exits with an error.
  • Generated page quality depends on source structure: clearer sections → prettier page.
  • Empty sections (highlights / features / roadmap) are hidden — no fake content.
  • If the user does not provide a skill path, ask them; do not auto-guess.

Dependencies

  • Python 3.8+ (stdlib only; no pip install required)
  • mistune v3.x — BSD-3-Clause, vendored unmodified at vendor/mistune/. If you delete vendor/, the script automatically falls back to a built-in regex renderer (simpler output, no crash).

Environment Variables

EnvDefaultPurpose
SKILL_INTRO_AUTHOR(unset)Override maintainer name (highest priority, before git config / $USER)
SKILL_INTRO_CACHE~/.cache/skill-introduction/cache.jsonCache file location (skill-name → dashboardId)
XDG_CACHE_HOME~/.cacheXDG base for cache when SKILL_INTRO_CACHE is unset
SKILL_INTRO_DEPLOY_CMD(unset)Path to deploy hook executable. ⚠️ Treated as trusted; only point at scripts you control.
SKILL_INTRO_UPDATE_ID(set by skill)Passed to deploy hook when updating an existing page

Third-party Notice

  • vendor/mistune/ — mistune v3.x, BSD-3-Clause License, © Hsiaoming Yang. Full license text: vendor/mistune/LICENSE. Summary: NOTICE.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.