agentleFS
vosslabvosslab/vosslab-skills/skills/documentation/readme-docs/SKILL.md

readme-docs

Skill5 starsChanged 21 days agoMITView on GitHub

Create or refresh a distinctive, evidence-backed `README.md` landing page for newcomers. Use when a README is missing, thin, generic, stale, or hard to navigate. Covers purpose, audience, first success, proof, examples, and documentation routes.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 5 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

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

    Start from assets/README_TEMPLATE.md for a missing, thin, or structurally failed README.

    line 16
  2. Very specific:

    Read references/readme_best_practices.md when project type or section choice is unclear.

    line 18
  3. Very specific:

    Read references/landing_page_ideas.md when the page needs a stronger signature promise or proof treatment.

    line 20
  4. Distinctive:

    Use one H1 for the project name, followed by a plain-prose opening paragraph of at most 250 characters.

    line 31
  5. Distinctive:

    Read operative repository rules, the current README, core docs, manifests, executable front doors, examples, tests, screenshots, …

    line 45
  6. Distinctive:

    Write a project-specific opening and choose a live demo, screenshot, output sample, compact transformation, architecture visual, or worked …

    line 52
  7. Distinctive:

    Verify every command that is safe and practical.

    line 54
  8. Very specific:

    When the project has a visual interface/output or the user requests screenshots, include this exact empty block for screenshot-docs:

    line 61
  9. Distinctive:

    Keep the value proposition, primary onboarding path, representative example, and navigation context in README.

    line 77

≈ “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

README landing page

Purpose

Make README.md a useful front door: establish identity, audience, value, proof, first success, and deeper documentation routes in the order a newcomer needs them. This skill owns README content; specialized documentation skills own their files under docs/.

Route supporting resources

  • Start from assets/README_TEMPLATE.md for a missing, thin, or structurally failed README.
  • Read references/readme_best_practices.md when project type or section choice is unclear.
  • Read references/landing_page_ideas.md when the page needs a stronger signature promise or proof treatment.
  • Apply references/review_checklist.md for broad refreshes and final landing-page audits.
  • Apply references/scoring_rubric.md for explicit scoring, before/after comparison, or a broad refresh.

Resolve every template placeholder before delivery.

Landing-page contract

  • Use one H1 for the project name, followed by a plain-prose opening paragraph of at most 250 characters. State purpose, audience or use case, main benefit, and a distinguishing detail when space allows. Put links, badges, code, URLs, and setup after that paragraph.
  • Establish purpose and the strongest available proof before Quick start.
  • Give a verified install/run path that reaches a meaningful result, plus one representative example and expected behavior when useful.
  • Summarize three to six concrete capabilities or outcomes and curate three to eight primary documentation routes before grouping a larger docset.
  • Include supported status, limitations, prerequisites, platforms, help, license, citation, or contributor routes only when repository evidence makes them useful.
  • Preserve current explanations, examples, visuals, warnings, and credits that remain accurate.

Workflow

  1. Read operative repository rules, the current README, core docs, manifests, executable front doors, examples, tests, screenshots, release/deployment evidence, and license files.
  2. Derive the project name, audience, primary value, project type, first-success path, strongest proof, and adoption constraints from current evidence.
  3. For a broad refresh, score the current page and identify the two highest-value improvements.
  4. Design the newcomer journey: identity and value, proof, first success, representative use, deeper routes, then secondary adoption context.
  5. Write a project-specific opening and choose a live demo, screenshot, output sample, compact transformation, architecture visual, or worked scenario that proves the signature promise.
  6. Verify every command that is safe and practical. Match paths, claims, live links, deployment URLs, status, and license wording to repository evidence.
  7. Apply the routed review checklist, repository Markdown/ASCII checks, and tests/test_readme_first_paragraph.py when present. Re-score broad refreshes.

Managed visuals

When the project has a visual interface/output or the user requests screenshots, include this exact empty block for screenshot-docs:

<!-- screenshots:begin (managed by screenshot-docs) -->
<!-- screenshots:end -->

This skill owns surrounding prose and the sentinel lines. screenshot-docs owns capture, docs/screenshots/, alt text, embed syntax, and the content between sentinels.

For a browser app, place a confirmed live link near the opening. Use a configured URL; derive a Pages URL only when deployment evidence confirms it.

Content boundary

Keep the value proposition, primary onboarding path, representative example, and navigation context in README. Route exhaustive APIs, configuration, architecture, troubleshooting, and maintainer process to their owning docs. Turn substantive missing docs and remaining landing-page opportunities into bounded follow-ups with an owner, target, evidence, success criteria, and verification.

Completion

Return the updated README, commands and links checked, unresolved claim status, and routed doc gaps. For broad refreshes, include before/after scoring and the strongest gain. Finish only when the quick start reaches a real result, the page carries project-specific proof when applicable, all placeholders are resolved, and current valuable content remains discoverable.

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.