agentleFS
ek0212ek0212/agent-ready-repo-kit/skills/technical-writing/SKILL.md

technical-writing

Skill0 starsChanged 4 months agoMITView on GitHub

Draft, revise, structure, and verify reader-centered technical documents that help people act, decide, understand a process, or use a system. Use when user asks for technical writing, documentation, instructions, SOPs, runbooks, manuals, technical descriptions or definitions, incident or progress reports, project proposals, recommendation reports, executive summaries, plain-language summaries, or technical workplace correspondence.

142 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use ketchup-video, principle-explain-the-number 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.
  • tristanphvnSkill

    ketchup-video

    Catch the user up like /keel:ketchup, with a short silent video of slides for what happened and the waiting calls as text under it. Use for /keel:ketchup-video.

    Also in 14 projects that use this

  • ianclearySkill

    principle-explain-the-number

    Apply before you trust, report, or act on a number you measured: a speedup, a regression, a throughput, a latency, or an eval result. Find what limits it, and rule out that it measured something other than the work you think.

    Also in 29 projects that use this

  • eryzerzSkill

    make-bot-ui

    Use when building a custom UI (page, dashboard, buttons) that should wake a Grok Bot over a webhook, when the user must provide a webhook sender key, or when exposing that UI on Tailscale.

    Also in 20 projects that use this

  • jaydubya818Skill

    principle-attack-the-premise

    Apply when repeated fixes fail and a shared assumption may be wrong.

    Also in 35 projects that use this

  • jaydubya818Skill

    principle-test-behavior-not-implementation

    Apply when tests mirror internals instead of asserting observable outcomes.

    Also in 36 projects that use this

  • ianclearySkill

    poteto-help

    Guides users through pstack setup, /poteto-mode, and picking the skill, playbook, or principle for a task. Type /poteto-help with a question.

    Also in 14 projects that use this

  • +5 more from jaydubya818

The file

Show its rules and text

We split this skill into the rules it gives an agent

26 rules. 0 are common, 26 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 134 edited versions we read keep these.

  • 26

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create technical documents that are clear, coherent, concise, concrete, correct, complete, and courteous.

    line 9
  2. Distinctive:

    Use for task-oriented or decision-oriented documents, including software and engineering documentation, instructions, SOPs, runbooks, …

    line 13
  3. Distinctive:

    Do not use for:

    line 15
  4. Distinctive:

    Document type, channel, length, template, and style requirements.

    line 27
  5. Distinctive:

    Ask one focused question only when a missing fact changes safety, scope, document type, or the required outcome.

    line 32
  6. Distinctive:

    Write a private one-line brief:

    line 38
  7. Distinctive:

    Before drafting, order the material around reader use:

    line 60
  8. Distinctive:

    Put the main message where readers need it, usually early.

    line 72
  9. Distinctive:

    Use active voice when the actor matters.

    line 74
  10. Distinctive:

    Use direct commands for procedures.

    line 75
  11. Distinctive:

    Keep one primary action per numbered step.

    line 76
  12. Distinctive:

    Use parallel wording for headings, steps, and list items.

    line 77
  13. Distinctive:

    Prefer concrete nouns, precise verbs, measured quantities, and defined conditions.

    line 78
  14. Distinctive:

    Use purposeful repetition when consistency matters more than variety.

    line 80
  15. Distinctive:

    Keep tone professional, respectful, inclusive, and proportionate to the evidence.

    line 81
  16. Distinctive:

    Check evidence for accuracy, sufficiency, relevance, completeness, currency, and authority.

    line 90
  17. Distinctive:

    Use facts and observations only in incident reports.

    line 92
  18. Distinctive:

    Use descriptive headings that expose the document structure.

    line 96
  19. Distinctive:

    Use numbered lists for sequence or priority.

    line 97
  20. Distinctive:

    Use tables for exact values or comparison.

    line 98
  21. Distinctive:

    Keep related elements together and preserve enough white space for scanning.

    line 101
  22. Distinctive:

    Never rely on color alone to convey meaning.

    line 102
  23. Distinctive:

    Run the checks that fit the document:

    line 106
  24. Distinctive:

    Use the full workflow for new, multi-section, high-stakes, safety-relevant, or evidence-heavy documents.

    line 122
  25. Distinctive:

    If evidence or requirements are incomplete, produce a labeled outline, questions, or placeholders.

    line 126
  26. Distinctive:

    If a mandated template, regulator, publisher, or organization standard conflicts with this skill, follow the authoritative specification …

    line 128

“Common” is measured: kept by at least 6 in 10 of the 134 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

Technical Writing

Create technical documents that are clear, coherent, concise, concrete, correct, complete, and courteous.

Scope

Use for task-oriented or decision-oriented documents, including software and engineering documentation, instructions, SOPs, runbooks, manuals, technical descriptions, definitions, reports, proposals, summaries, memos, and technical email.

Do not use for:

  • Explanation-only requests with no document deliverable. Use clear-explanation.
  • Narrative articles, profiles, or essays. Use creative-writing.
  • Literature reviews or research synthesis. Use stanford-research.
  • AI-pattern cleanup when structure and technical usability are already sound. Use no-ai-slop.
  • Spoken presentation decks. Use presentation-builder.

Inputs

  • Audience, expertise, needs, and likely objections.
  • Purpose and action or decision the document must support.
  • Document type, channel, length, template, and style requirements.
  • Verified facts, evidence, source material, terminology, and constraints.
  • Safety, legal, privacy, accessibility, or operational requirements.
  • Existing draft and required sections, when revising.

Ask one focused question only when a missing fact changes safety, scope, document type, or the required outcome. Otherwise state a narrow assumption.

Workflow

Step 1: Frame The Rhetorical Situation

Write a private one-line brief:

For [audience], this [document type] will help them [task or decision] in [context], using [required evidence or constraints].

Identify:

  • What readers already know.
  • What they need before they can act.
  • What success looks like after reading.
  • Which terms need definition.
  • Which concerns, risks, or resistance need coverage.

Step 2: Choose The Document Pattern

Select the smallest document type that fits the purpose. Follow an existing template or required specification before a generic pattern.

Load references/document-types.md when the request names a genre or the structure is uncertain.

Step 3: Map The Information

Before drafting, order the material around reader use:

  1. Main point, task, recommendation, or expected outcome.
  2. Context and prerequisites.
  3. Steps, parts, evidence, or findings.
  4. Risks, limits, alternatives, and troubleshooting.
  5. Next action or decision.

Separate verified facts, interpretations, assumptions, and unknowns. Do not invent missing specifications, results, quotes, incidents, costs, or citations.

Step 4: Draft For Reader Action

  • Put the main message where readers need it, usually early.
  • Give each paragraph one central idea and a clear topic sentence.
  • Use active voice when the actor matters.
  • Use direct commands for procedures.
  • Keep one primary action per numbered step.
  • Use parallel wording for headings, steps, and list items.
  • Prefer concrete nouns, precise verbs, measured quantities, and defined conditions.
  • Define necessary jargon on first use. Keep established terms for specialist readers.
  • Use purposeful repetition when consistency matters more than variety.
  • Keep tone professional, respectful, inclusive, and proportionate to the evidence.
  • Qualify claims when evidence is limited. Avoid unsupported certainty.

For persuasive documents, connect each claim to a reason and credible evidence. Address material counterarguments fairly.

Step 5: Integrate Evidence

  • Cite quoted, paraphrased, summarized, and adapted material.
  • Explain how each source or data point supports the surrounding claim.
  • Check evidence for accuracy, sufficiency, relevance, completeness, currency, and authority.
  • Preserve exact wording only when quotation is necessary.
  • Use facts and observations only in incident reports. Label unresolved information.

Step 6: Design For Navigation

  • Use descriptive headings that expose the document structure.
  • Use numbered lists for sequence or priority. Use bullets for unordered items.
  • Use tables for exact values or comparison. Use figures for patterns, processes, or spatial relationships.
  • Introduce each visual in the text, place it near the relevant passage, and add an informative label, caption, source, and accessible text alternative.
  • Apply contrast, alignment, repetition, proximity, and size consistently.
  • Keep related elements together and preserve enough white space for scanning.
  • Never rely on color alone to convey meaning.

Step 7: Verify Through The Reader's Task

Run the checks that fit the document:

  • Can the named audience find the main point quickly?
  • Can a reader complete the task or make the decision without hidden prerequisites?
  • Are steps complete, ordered, testable, and safe?
  • Are warnings placed before the related hazard and repeated at the relevant step?
  • Are claims supported and limitations visible?
  • Are headings, lists, tables, figures, labels, units, and terms consistent?
  • Are source uses cited and traceable?
  • Does every section serve the stated purpose?
  • Is the document free of ambiguity, bias, avoidable jargon, grammar errors, and formatting drift?

For instructions, simulate the procedure or have a representative user test it when practical. Record observed failure points and revise.

Applicability And Fallbacks

Use the full workflow for new, multi-section, high-stakes, safety-relevant, or evidence-heavy documents.

For a short low-risk edit, preserve the existing structure and run Steps 1, 4, and 7 only.

If evidence or requirements are incomplete, produce a labeled outline, questions, or placeholders. Do not turn assumptions into facts.

If a mandated template, regulator, publisher, or organization standard conflicts with this skill, follow the authoritative specification and note the conflict.

Output

Return the requested document first.

Then include only what the user needs to finish it:

  • Assumptions.
  • Evidence or source gaps.
  • Required verification or user testing.
  • Alternate structure only when it materially changes usability.

Source

This skill adapts a human-authored technical writing handbook. Read references/source-and-license.md for attribution and license terms.

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.