Index of every page on this site: llms.txt
All skills
harlan-zw avatar

/im-part-human

@b80e2f4
Run it yourself

Edits prose that people read, such as blog posts, docs, social posts, and email, for clear meaning, accurate claims, concise wording, and the author's own voice. Use for general prose editing, tightening, or humanizing, including explicitly requested wording-only edits. Release-note creation and substantive revision belong to release-notes.

Use this Skill: https://skilld.dev/gh/harlan-zw/brundlefly/im-part-human

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencesreader-attention.md

≈766 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Allocate the reader's attention

Identify what the reader came to decide, understand, or do. Reader attention is finite. Predict how much attention this piece will receive before choosing its scope and depth. Use supplied preferences; otherwise state a reasonable assumption in private notes. Treat that estimate as an editorial choice, not a measured reading time.

Review the whole document before polishing individual sentences. A document can use plain words and short paragraphs while demanding too much attention.

Decide what earns space

Give space according to consequence for this reader, rather than the number of changes or the author's effort.

  • Lead with the facts that affect the reader's decision or next action.
  • Explain important changes enough to show their effect, evidence, and limits.
  • Give smaller changes a brief mention or a useful reference.
  • Put specialist detail beside its use, in a linked guide, or in an optional reference section.

Do not give every feature the same heading, paragraph count, example, or explanation of why it matters. Do not fill a template's available slots with weaker points. Shorten low-value explanations before compressing the important explanation until it becomes cryptic. Avoid turning everything into a dense list of nouns and identifiers.

Preserve critical prerequisites, warnings, and actions on the main reading path. Optional detail must not hide a condition the reader needs before acting. Keep substantive meaning through compression or a checked reference. If removing meaning falls outside the requested scope, propose that removal before editing it away.

Apply to release notes

An initial scan should help the reader decide whether to upgrade and identify any required changes. Give new capabilities and measured improvements enough detail to assess their value. Lead with the measured gains that convey the most useful meaning to this reader. Keep the baseline and essential qualifications beside them. Link to raw results and method for detail. Do not choose a weaker baseline to make a larger percentage or imply speed from size alone. Make support requirements and consequential migration warnings easy to find. Keep smaller fixes concise. Use the changelog and migration guide for complete lookup detail.

Use a different shape for each destination when it serves the reader. A GitHub release can introduce the changes and link to a longer article. Do not duplicate that whole article by default.

Check the reading path

Read only the opening, headings, and first lines. What can the reader decide from that scan? Then read the main path without optional detail. Does it provide enough context to act safely? Check whether repeated explanations add meaning or just require another pass over the same fact. Check that important changes receive more attention than minor fixes.

Record concrete reading problems and repairs outside the published text. Word counts, punctuation checks, and an agent's approval cannot establish that a human reader finds the result useful.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts1mo3 checks · Risk SAFE
  • Gen Agent Trust Hub1mo

    The skill is a comprehensive set of editorial guidelines for improving prose clarity and accessibility. It consists entirely of Markdown instructions and reference materials, containing no executable code, tool definitions, or network capabilities.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: LOW · No issues

Signed by skilld at b80e2f4. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Activeupdated last month

README badge

README badge for harlan-zw/brundlefly/im-part-human