agentleFS
durchnulldurchnull/render/skills/init/SKILL.md

init

Skill0 starsChanged 3 months agoMITView on GitHub

Scaffold the render config (.render/ with config.py + pages/) for this project, or refresh an existing one against the installed plugin

137 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use daily-arxiv, exp-eval and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • LambenthanSkill

    daily-arxiv

    每日从 arXiv 拉取新论文,过滤相关性,auto-ingest 高优先级论文,检测 SOTA 更新

    Also in 5 projects that use this

  • LambenthanSkill

    exp-eval

    实验判决门:Review LLM 独立评判实验结果 → 4 种判决路径 → 自动更新 claims confidence、ideas status、graph edges

    Also in 5 projects that use this

  • skyllwtSkill

    exp-pilot-eval

    Pilot result evaluation — read pilot results, apply success criteria, update idea page (pilot_result, failure_reason if failed), generate PILOT_VERDICT_REPORT. Called by /ideate Phase 5 after /exp-pilot-run.

    Also in 4 projects that use this

  • skyllwtSkill

    exp-pilot-run

    Pilot experiment execution — read Pilot Spec YAML, write pilot code, run experiment(Confirm with the user before operation and require the applicant to conduct manual inspection), return results. Called by /ideate Phase 5. Does NOT modify wiki pages or judge pass/fail.

    Also in 4 projects that use this

  • gaebalaiSkill

    codex-system

    PROACTIVELY consult Codex CLI, your highly capable supporter with exceptional reasoning and task completion abilities. Codex is a trusted expert you should ALWAYS consult BEFORE making decisions on: design choices, implementation approaches, debugging strategies, refactoring plans, or any non-trivial problem. When uncertain, consult Codex. Don't hesitate - Codex provides better analysis. Explicit triggers: "think deeper", "analyze", "second opinion", "consult codex".

    Also in 3 projects that use this

  • +6 more from Lambenthan
  • +1 more from skyllwt

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 125 edited versions we read keep these.

  • 0

    Distinctive

    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:

    If a qualifying config already exists (.render/ with both config.py and pages/__init__.py), this is a refresh, not a scaffold — the command …

    line 9
  2. Very specific:

    Ask the user which pages the project needs if the request doesn't say.

    line 23

“Common” is measured: kept by at least 6 in 10 of the 125 edited versions of this file we could read. “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

Set up the render plugin for the current project:

  1. If a qualifying config already exists (.render/ with both config.py and pages/__init__.py), this is a refresh, not a scaffold — the command a project runs after updating the plugin:

    python3 "${CLAUDE_PLUGIN_ROOT}/engine/scaffold.py"

    It replaces the four files the plugin owns (engine_locator.py, render.py, README.md, pages/__init__.py) whenever they differ from the installation, and creates any that are missing. It never touches config.py, content.py, kinds/ or anything under pages/<id>/ — the project's own design is not yours to overwrite, then or ever. Report its output: which files were refreshed, and that the project's files were left alone. Pass --check first when the user wants to see what would change without changing it. Then stop — a configured project needs nothing else from this skill.

  2. Otherwise scaffold it in one command — never copy the files by hand:

    python3 "${CLAUDE_PLUGIN_ROOT}/engine/scaffold.py" --fresh

    It writes the full template scaffold (config.py, content.py, pages/__init__.py, the example dashboard page, render.py, engine_locator.py, README.md) and never overwrites a file that already exists. A fresh copy is by definition current — the engine compares bytes, so a new directory needs nothing further to be recognised as up to date.

  3. Ask the user which pages the project needs if the request doesn't say. The example dashboard page is a starter, not an assumption — rename it, redesign it or replace it entirely; every page is the project's own design.
  4. For each page, scaffold the folder instead of typing it:

    python3 "${CLAUDE_PLUGIN_ROOT}/engine/scaffold.py" --new-page <id> --section <sid> --section <sid> --title "…"

    Then make it real: each section stub's build() raises until it renders something, and DESCRIPTION is empty until you write the one sentence the index card shows — --check reports a section page without one. Each page renders to output/<id>.html (override with FILENAME; move all output with OUT_DIR in config.py). A page that should render one output per data file takes --kind <k> --sources '<glob>' instead of --section — see the new skill, or /render:questionnaire and /render:checklist for the two kinds that ship.

  5. Read ${CLAUDE_PLUGIN_ROOT}/design-manual.md (the core: principles, tokens, hierarchy, component catalog) before writing any section markup — colors, typography and components come exclusively from the engine's design_system.py (TOKENS + BASE_CSS); no custom hex values. Then pick the components from the catalog in its §6 and read the listed docs/design/ reference file for every helper you call — the catalog is an index, not a spec.
  6. Render and verify:
    python3 "${CLAUDE_PLUGIN_ROOT}/engine/render.py" --all
    python3 "${CLAUDE_PLUGIN_ROOT}/engine/render.py" --check
    (Equivalent once scaffolded: python3 .render/render.py --all / --check — the thin entry point resolves the engine itself.)
  7. Tell the user where the pages land (output/ next to the config unless OUT_DIR says otherwise), that output/index.html is the entry point the engine keeps up to date on its own — a card per page, linking to it — and that the plugin's PostToolUse hook now re-renders all of it automatically after every Write/Edit. Mention once that running this command again after a plugin update refreshes the plugin's own files in .render/ and leaves theirs untouched; the engine says so itself when they fall behind.

Arguments (optional): $ARGUMENTS — if a directory is given, use it instead of <project>/.render/.

The extension contract

Everything project-specific lives in .render/ — config, pages, content helpers, and any standalone page scripts the project adds later. The engine is imported from the plugin installation, never copied: engine_locator.py resolves it, and page_api is the only stable interface for project-local code.

Four files in that directory are the plugin's rather than the project's — engine_locator.py, render.py, README.md, pages/__init__.py — and each says so in its own header. They are the only things a plugin update leaves behind, which is why re-running this command is all a project ever has to do; engine/scaffold.py owns that split, and nothing else may decide it case by case. When the project needs another page (a checklist, a form/questionnaire, a report, a second dashboard), follow the new skill — a page is a folder under pages/, and the engine picks it up on its own.

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.