voice
Skill2 starsChanged 25 days agoApache-2.0View on GitHub
Load for project-setup voice capture, every drift_interval chapters, or when the author says the prose does not sound like them. The voice skill owns the voice lifecycle: capture, measured profile, exemplars/anti-exemplars, drift checks, the blind-tag falsification test, and retune. It is the strongest signal the style-creator and critics have — and the reference the mechanical net defers to.
35 other projects use a version of this.
Adopted by
Most recently taken up by
- Deupaxx/EveryDay-Writertheir own version, first seen Nov 16, 2026
- ShiroTsuma/Voice-MCPtheir own version, first seen Nov 15, 2026
- igoroliveirg/claude-copytheir own version, first seen Nov 15, 2026
- codeboehm/maximtheir own version, first seen Nov 15, 2026
- guaardvark/guaardvarktheir own version, first seen Nov 13, 2026
- BermanRoman/promt-seo-geotheir own version, first seen Nov 8, 2026
Used with
Projects with it also use claude-buddy, face-enroll 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.
autonomous-aiSkillclaude-buddy
Coordinate with the Claude Buddy companion (Claude Desktop + Claude Code on the user's Mac) — voice-approve tool permission prompts and stay aware of Claude activity.
Also in 3 projects that use this
autonomous-aiSkillface-enroll
Manage the device's face recognition roster — enroll new faces (3 paths: user-supplied photo, agent-captured snapshot on user request, or HAL's familiar-stranger prompt) and maintain the enrolled set (status / remove / reset). All enrolled persons are friends; strangers stay unnamed until promoted via one of the enroll flows.
Also in 3 projects that use this
- +10 more from autonomous-ai
Other versions35 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 35
- autonomous-ai/autonomous-os★ 569same name
- autonomous-ai/lampOS★ 569same name
- autonomous-ai/Physical-AI-Operating-System★ 569same name
- guaardvark/guaardvark★ 320same name
- jvm-skills/jvm-skills★ 140same name
- simota/agent-skills★ 99same name
- crisandrews/ClawCode★ 64same name
- Deupaxx/EveryDay-Writer★ 39same name
- ggombee/code-forge★ 18same name
- shreyaskarnik/voice-mcp★ 17same name
25 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository33 other files it gives its agents.
- character-simSkill
- cold-readSkill
- creative-researchSkill
- creative-writing-craftSkill
- creative-writing-modesSkill
- creative-writing-museSkill
- demolitionSkill
- exportSkill
- gatesSkill
- grill-with-docsSkill
- information-hierarchySkill
- intent-modelingSkill
- kb-integritySkill
- kb-managementSkill
- knowledge-layersSkill
- llm-writingSkill
- md-validationSkill
- project-setupSkill
- qi-layerSkill
- reader-simSkill
- reflectSkill
- series-bibleSkill
- seriesSkill
- shared-daoSkill
- story-ledgersSkill
- story-memorySkill
- story-planningSkill
- story-reviewSkill
- structured-artifactSkill
- style-guardrailsSkill
- writing-principlesSkill
- writing-staffingSkill
- zoom-outSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 27 edited versions we read keep these.
0
Distinctive
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
- Very specific: line 30
Drift check: every drift_interval chapters or on author request: mechanical + LLM passes.
“Common” is measured: kept by at least 6 in 10 of the 27 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.
Files
Voice
The voice skill owns the full lifecycle of a project's prose voice: how it is captured from the author, measured into a numeric profile, calibrated against exemplar passages, checked for drift, tested for whether the capture worked at all, and retuned when the author says it is off. It is the reference the mechanical net (style-guardrails/) defers to via the measured-profile law, and the primary lens the style-creator and critics use.
What this skill is
- The single source of truth for the project's voice:
kb/styles/voice.md(the document these resources describe how to build and maintain). - The measured-profile law made numeric:
kb/styles/baseline.mdis the author's own per-1k rates, and the net measures against them instead of a generic average. - A self-contained reference: it restates the measured-profile law, the drift/retune loop, and the blind-tag test so the style-creator can act without loading other skills. Cross-references to
style-guardrails/andcreative-writing-craft/are pointers only.
Load the resource needed
resources/voice-template.md— the two-partkb/styles/voice.mdstructure: Part 1 guardrails (restated inline) and Part 2 identity scaffold (tone, sentence rhythm, vocabulary register, POV/tense, dialogue conventions, exemplar passages, anti-exemplars). How it is built, used, and maintained.resources/voice-profile.md— the measured per-1k voice profile: what it is, howvellum style stats --baselineemits it, the bidirectional diff (over-shoot vs. under-shoot), how it overrides the tier bans, how the baseline is built and refreshed, and the blind-tag test that validates the capture.resources/blind-tag-test.md— the falsifiable check that voice capture worked: shuffle author-corpus and manuscript passages, strip tags, sort with a fresh project-naive critic; < 75% correct attribution means the capture failed and the exemplars need redoing.resources/character-dialogue-profiles.md— the per-character analog of the blind-tag test: the optional character-pagedialogue profilesection (idiolect markers, preferred deflections, sentence-length signature, top tokens) built from@character-simat page creation, and the ~70% blind speaker-attribution test on attribution-stripped dialogue lines; the mechanical companion isvellum style stats --dialogue(per-speaker stats + pairwise convergence flags, report-only).resources/drift-check.md— the two-pass drift procedure (mechanicalstyle stats --baseline+ LLM exemplar/anti-exemplar comparison), when to run it, and how to resolve the two passes.resources/retune.md— the voice realignment protocol: take the author's unguided sample, analyze the gap, one alignment pass (no structural changes, no quality evaluation of the sample), present, and harvest the sample back into the voice profile.
How the voice lifecycle fits the chapter loop
- Setup (
project-setup): collect 3–5 author samples → style-creator buildskb/styles/voice.md+kb/styles/baseline.md→ run the blind-tag test once to confirm the capture. - Drafting: writer loads voice Part 1 + Part 2 + one exemplar (§9 recipe slot 6). Mechanical net (
style-guardrails/) runs silently, defers to the measured profile. - Drift check every
drift_intervalchapters or on author request: mechanical + LLM passes. Clean → continue. Drift → retune. - Retune on "this doesn't sound like me": author sample → one alignment pass → harvest into voice.md + baseline.md → drift check to confirm.
- Author-edit harvesting (style-creator, ongoing): diff author edits against agent drafts, promote characteristic choices into exemplars — the strongest available voice signal, free.
- Milestone: critics judge execution against exemplars/anti-exemplars; blind-tag test re-run after any major retune.
The measured-profile law (restated)
The author's own measured rates are the standard. style-guardrails/ states the law; this skill provides the numbers. Caps adapt to kb/styles/baseline.md; only statistically significant over-shoot vs. the author's baseline is a finding; under-shoot is reported too. A word the author genuinely uses is never slop in this book.
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
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.

