agentleFS
timscheueraitimscheuerai/gtm-architecture-template/.agents/skills/voice-calibration/SKILL.md

voice-calibration

Skill2 starsChanged 2 months agoMITView on GitHub

Build or refine an author's writing guide from their own writing samples and edits, without importing another creator's voice.

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

7 rules. About 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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. Very specific:

    Read AGENTS.md, voice/linkedin-voice.md, relevant identity and audience pages, and the source samples.

    line 11
  2. Distinctive:

    Use several representative samples when available.

    line 17
  3. Distinctive:

    Look for recurring choices in opening lines, sentence rhythm, paragraph length, vocabulary, punctuation, specificity, argument structure …

    line 19
  4. Very specific:

    Write observed rules in voice/linkedin-voice.md and format choices in voice/formats.md.

    line 21
  5. Distinctive:

    When drafting is requested, use a supplied, public-use source to write a short sample.

    line 25
  6. Distinctive:

    If writing samples are missing, collect them or produce a clearly labeled neutral draft when that is useful and requested.

    line 27
  7. Very specific:

    Update source notes, index and log.

    line 29

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Workspace: all context paths below (identity, audience, strategy, voice, brand, raw, output, index, log, workspace preferences and context scripts) resolve from context/ in the kit root. Read both the root AGENTS.md and context/AGENTS.md. Skill-local references/ and flowchart scripts/ resolve from this skill directory.

Calibrate the voice

Read AGENTS.md, voice/linkedin-voice.md, relevant identity and audience pages, and the source samples. Capture new sources through capture-context when needed.

Separate writing authored by the person, transcripts of speech, approved ghostwritten posts, rejected drafts and external inspiration. Human edits show preferences; engagement numbers show outcomes, not who wrote a sentence.

Derive rules from examples

Use several representative samples when available. One sample supports tentative observations, not a universal style. Record the limits of the evidence.

Look for recurring choices in opening lines, sentence rhythm, paragraph length, vocabulary, punctuation, specificity, argument structure and endings. Quote short illustrative passages with source paths. Preserve meaningful irregularities instead of making the voice more polished by default.

Write observed rules in voice/linkedin-voice.md and format choices in voice/formats.md. Distinguish consistent patterns from context-dependent choices. Do not hard-code a dialect, numeric separator, slang quota, hook length or CTA frequency from this template.

Calibrate with an actual draft

When drafting is requested, use a supplied, public-use source to write a short sample. Give the author a concrete passage to edit. File their actual corrections as evidence, then update the few rules the corrections support. Do not label your own imitation approved.

If writing samples are missing, collect them or produce a clearly labeled neutral draft when that is useful and requested. Do not claim to reproduce a specific voice from a biography alone.

Update source notes, index and log; run scripts/wiki-lint.sh. Return the guide and any unresolved voice choices.

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.