agentleFS
matheussluzzmatheussluzz/ZeuZ-Agent/skills/hermes/SKILL.md

hermes

Skill0 starsChanged 4 months agoMITView on GitHub

Translate complex technical, data, product, legal, or research material into plain, commercially useful language without changing facts, uncertainty, obligations, numbers, or limitations. Use for executive summaries, stakeholder explanations, business-facing documentation, onboarding, or when the user is unfamiliar with the subject.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Prefer active voice, short sentences, informative headings, descriptive links, and concrete actors.

    line 30
  2. Very specific:

    Run node scripts/check-invariants.mjs <source> <translation> as a conservative lexical backstop.

    line 38
  3. Distinctive:

    Read references/plain-language-checklist.md for risk tiers, channel checks, examples, and the semantic verification matrix.

    line 42

≈ “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

Hermes

Translate for a real audience and decision. Do not merely shorten the text, infantilize the reader, or turn evidence into marketing.

Progressive workflow

Load only the stage needed next. Preserve the compact outputs from completed stages instead of repeatedly loading the whole source.

1. Freeze the communication brief

Establish audience, demonstrated prior knowledge, channel, desired decision/action, time available, tone, and source authority. Use assets/communication-brief.md. Ask only when a missing choice changes the result. Do not silently optimize for an imagined executive.

Gate: name one primary audience and one intended outcome. If there are materially different audiences, create separate versions.

2. Build the meaning contract

Extract the source's claims and invariants: numbers, units, currencies, dates, entities, scope, negations, obligations, conditions, exceptions, uncertainty, confidence, citations, and causal language. Classify each statement as source fact, source opinion, or your inference. Keep sensitive details out of examples.

Gate: resolve contradictory passages or show the contradiction; do not choose the convenient version.

3. Translate in layers

Lead with the answer and why it matters. Follow with confirmed impact, required decision/action, risks and limitations, next step, and a small glossary when useful. Define the exact technical term on first use and expand non-trivial acronyms. Use an example, comparison, or diagram only when it preserves the same mechanism and boundary conditions.

Prefer active voice, short sentences, informative headings, descriptive links, and concrete actors. Keep the original legal, regulatory, contractual, or safety-critical language available beside the explanation; the translation never replaces it.

4. Challenge commercial claims

Mark inferred business impact as inference. Never invent return on investment, savings, causality, urgency, certainty, endorsement, or a recommendation. Preserve must versus should, may versus will, correlation versus causation, and estimate versus commitment.

5. Verify meaning and usability

Run node scripts/check-invariants.mjs <source> <translation> as a conservative lexical backstop. Then trace every meaning-contract item into the translation. For high-impact material, ask a representative reader to paraphrase the meaning and intended action without coaching; repair the message and repeat.

Delivery evidence: include the audience/outcome, translated artifact, important terms retained, assumptions/inferences, unresolved ambiguity, and verification performed. A readability score or passing invariant script does not prove comprehension or semantic equivalence.

Read references/plain-language-checklist.md for risk tiers, channel checks, examples, and the semantic verification matrix.

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.