agentleFS
baoyuboooobaoyuboooo/simple-agent-skills/skills/simple-doc/SKILL.md

simple-doc

Skill0 starsChanged 5 months agoMITView on GitHub

Use when simplifying or condensing documents that should become clearer, more direct, and easier to follow without changing confirmed meaning.

  • In this repositoryOne other file it gives its agents.

The file

Show its rules and text

We split this skill into the rules it gives an agent

42 rules. About 0 are common, 39 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈39

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If reliable judgment is not possible, ask a human instead of guessing.

    line 12
  2. Distinctive:

    Do not auto-trigger this skill on broad requests such as:

    line 26
  3. Distinctive:

    Default to suggestions only.

    line 43
  4. Distinctive:

    Do not directly modify the document unless the user explicitly asks for execution with wording such as:

    line 45
  5. Distinctive:

    If a request mixes simplification with non-simplification work, keep this skill scoped to the simplification portion only.

    line 57
  6. Distinctive:

    If the request can be split cleanly, simplify the relevant part and leave the non-simplification part to another workflow or human decision.

    line 59
  7. Distinctive:

    If the request cannot be split safely, ask the human to separate the simplification task from the rest.

    line 60
  8. Distinctive:

    If explicit invocation is used for a non-simplification task such as drafting a new document from scratch, do not treat that as valid just …

    line 61
  9. Distinctive:

    Do not rewrite unclear content as if it were confirmed.

    line 71
  10. Distinctive:

    If the task has a required template or explicitly required structure, preserve that structure before simplifying anything else.

    line 72
  11. Distinctive:

    Check whether the outline has a clear top-down shape.

    line 78
  12. Distinctive:

    Keep heading levels shallow.

    line 79
  13. Distinctive:

    Prefer fewer clear sections over deep nesting.

    line 80
  14. Distinctive:

    Keep entry documents lightweight.

    line 83
  15. Distinctive:

    Do not expand the document scope to make it feel complete.

    line 84
  16. Distinctive:

    Do not delete required sections or rewrite fixed structure merely to make the document feel simpler.

    line 85
  17. Distinctive:

    Every document and section should reveal the point before the details.

    line 89
  18. Distinctive:

    Put conclusions before explanations.

    line 91
  19. Distinctive:

    Put stable rules before examples.

    line 92
  20. Distinctive:

    Put required actions before optional notes.

    line 93
  21. Distinctive:

    Put open questions in a dedicated place instead of scattering uncertainty through the body.

    line 94
  22. Distinctive:

    Avoid long setup paragraphs before the actual rule, decision, or action.

    line 95
  23. Distinctive:

    When some inputs are still unresolved, improve structure and wording only within confirmed boundaries, and keep unresolved items explicit …

    line 103
  24. Distinctive:

    Every change must support the document goal.

    line 127
  25. Distinctive:

    Do not rewrite stable sections without a reason.

    line 128
  26. Distinctive:

    Do not introduce new facts, rules, or scope.

    line 129
  27. Distinctive:

    Do not silently change confirmed meaning.

    line 130
  28. Distinctive:

    Prefer small structural fixes before broad rewording.

    line 131
  29. Distinctive:

    Keep terminology consistent with the source material.

    line 132
  30. Distinctive:

    If structure constraints are unclear, avoid irreversible structural changes until the requirement is clearer, but still make safe local …

    line 133
  31. Distinctive:

    When document issues are found, provide simplification suggestions by default rather than editing immediately.

    line 137
  32. Distinctive:

    Keep the suggestions scoped to the current document goal.

    line 139
  33. Distinctive:

    When editing is explicitly requested, apply the smallest change that improves clarity without changing confirmed meaning.

    line 141
  34. Distinctive:

    Ask for human input when facts, scope, templates, acceptance criteria, confirmed meaning, or the correct structural or content decision …

    line 143
  35. Distinctive:

    Do not present the document as final when unresolved decisions remain.

    line 144
  36. Very specific:

    Use simple ordered lists for output items: 1.

    line 170
  37. Distinctive:

    Do not include empty sections.

    line 171
  38. Distinctive:

    Do not include reasoning-process narration.

    line 172
  39. Distinctive:

    Keep output scoped to simplification only.

    line 173
  40. Distinctive:

    Keep the response bounded to the completed task.

    line 174
  41. Very specific:

    Use ## 🧪 Verification to state what was checked and whether meaning, scope, confirmed facts, required structure, and constraints were …

    line 175
  42. Very specific:

    Use ## ❓ Open Questions for unresolved decisions, missing inputs, or unverified content that still needs human confirmation.

    line 176

≈ “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

Simple Doc

Make documents easier to understand, maintain, and execute through simplification. Do not merely make them shorter.

Simplification must not override user-stated goals, scope, and constraints, or required templates, rules, or fixed structure.

If reliable judgment is not possible, ask a human instead of guessing.

Trigger Boundary

Auto-trigger this skill only when the request shows explicit simplification intent.

Examples of valid intent:

  • simplify
  • condense
  • reduce redundancy
  • make this more direct
  • make this shorter by removing unnecessary content

Do not auto-trigger this skill on broad requests such as:

  • optimize
  • adjust
  • review
  • rewrite
  • design
  • organize

optimize only counts when paired with explicit simplification intent.

Explicit direct invocation such as $simple-doc remains valid even when the request does not include auto-trigger wording.

Explicit invocation does not widen this skill into general drafting, document design, or general rewrite work. The request must still be a simplification task.

Default Action

Default to suggestions only.

Do not directly modify the document unless the user explicitly asks for execution with wording such as:

  • auto-simplify
  • apply the suggestions
  • directly modify

These are representative examples, not a closed whitelist.

Explicit execution wording only enables modification within a simplification task. It does not convert this skill into a general authoring or drafting skill.

Mixed-Intent Requests

If a request mixes simplification with non-simplification work, keep this skill scoped to the simplification portion only.

  • If the request can be split cleanly, simplify the relevant part and leave the non-simplification part to another workflow or human decision.
  • If the request cannot be split safely, ask the human to separate the simplification task from the rest.
  • If explicit invocation is used for a non-simplification task such as drafting a new document from scratch, do not treat that as valid just because the skill name was invoked.

1. Think Before Editing

Understand the document goal before changing structure or wording.

  • Identify the document type, audience, and intended use.
  • Identify the simplification goal and what should become clearer, shorter, or more direct.
  • Identify confirmed facts, rules, scope, constraints, templates, paths, and acceptance criteria.
  • Identify unclear or unconfirmed content before changing wording or structure.
  • Do not rewrite unclear content as if it were confirmed.
  • If the task has a required template or explicitly required structure, preserve that structure before simplifying anything else.

2. Outline First

Inspect the document structure before editing sentences.

  • Check whether the outline has a clear top-down shape.
  • Keep heading levels shallow.
  • Prefer fewer clear sections over deep nesting.
  • Merge duplicated sections.
  • Split sections that mix background, rules, workflow, checklist, or examples.
  • Keep entry documents lightweight; point to deeper documents instead of copying full details.
  • Do not expand the document scope to make it feel complete.
  • Do not delete required sections or rewrite fixed structure merely to make the document feel simpler.

3. Top-Down Structure

Every document and section should reveal the point before the details.

  • Put conclusions before explanations.
  • Put stable rules before examples.
  • Put required actions before optional notes.
  • Put open questions in a dedicated place instead of scattering uncertainty through the body.
  • Avoid long setup paragraphs before the actual rule, decision, or action.

4. Final-State Content Only

Formal documents should contain final decisions, not the path used to reach them.

For formal specs, runbooks, policies, and published guidance, prefer final-state content only.

When some inputs are still unresolved, improve structure and wording only within confirmed boundaries, and keep unresolved items explicit as open questions instead of inventing final answers.

Remove or rewrite:

  • brainstorming residue
  • abandoned alternatives
  • "we can consider" language
  • excessive rationale
  • repeated explanations
  • vague goals with no executable meaning

Keep only:

  • confirmed conclusions
  • confirmed rules
  • confirmed workflow
  • confirmed scope
  • confirmed actions
  • necessary open questions

5. Goal-Driven Surgical Editing

Edit only what improves the document for its real purpose.

  • Every change must support the document goal.
  • Do not rewrite stable sections without a reason.
  • Do not introduce new facts, rules, or scope.
  • Do not silently change confirmed meaning.
  • Prefer small structural fixes before broad rewording.
  • Keep terminology consistent with the source material.
  • If structure constraints are unclear, avoid irreversible structural changes until the requirement is clearer, but still make safe local improvements where useful.

Suggestions-First Rule

When document issues are found, provide simplification suggestions by default rather than editing immediately.

  • Keep the suggestions scoped to the current document goal.
  • Apply edits only when the user explicitly requests modification.
  • When editing is explicitly requested, apply the smallest change that improves clarity without changing confirmed meaning.
  • Re-check the updated document as a whole, not just the edited passage: outline, section responsibility, placement, surrounding fit, final-state wording, and open questions.
  • Ask for human input when facts, scope, templates, acceptance criteria, confirmed meaning, or the correct structural or content decision cannot be determined reliably from the available context.
  • Do not present the document as final when unresolved decisions remain.

Output Contract

Return only the relevant sections.

For suggestion mode, use:

  • ## 📌 Main Verdict
  • ## 🟥 High-Priority Suggestions
  • ## 🟨 Medium-Priority Suggestions
  • ## 🟩 Low-Priority Suggestions
  • ## 🛡 Meaning-Preservation Notes
  • ## ❓ Open Questions

For auto-modification mode, use:

  • ## 📌 Main Verdict
  • ## 🛠 Changes Made
  • ## 🛡 Preservation Notes
  • ## 🟨 Remaining Concerns
  • ## 🧪 Verification
  • ## ❓ Open Questions

Rules:

  • Use simple ordered lists for output items: 1. 2. 3.
  • Do not include empty sections.
  • Do not include reasoning-process narration.
  • Keep output scoped to simplification only.
  • Keep the response bounded to the completed task. Do not add generic follow-up questions.
  • Use ## 🧪 Verification to state what was checked and whether meaning, scope, confirmed facts, required structure, and constraints were preserved.
  • Use ## ❓ Open Questions for unresolved decisions, missing inputs, or unverified content that still needs human confirmation.

Final Verification

  • The structure is clearer, or required structure was preserved correctly when present.
  • The document uses top-down structure.
  • Reasoning residue, repeated content, and vague wording are removed.
  • Confirmed facts, rules, scope, paths, templates, and acceptance criteria are preserved.
  • New content does not invent unsupported facts, rules, or scope.
  • Unclear inputs and unconfirmed content are listed under ## ❓ Open Questions.

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.