example-voice
Skill55 starsChanged 17 days agoMITView on GitHub
Apply the owner's writing voice to any draft. Use when a post, script, newsletter, or reply needs to sound like the owner instead of generic AI prose: tone, persona, sentence rhythm, anti-slop rules, and first-person grounding. This is a style layer, not a first-draft writer.
In this repository5 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- html-prototypesSkill
- scheduled-runsSkill
- vault-maintenanceSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
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
- Distinctive: line 24
If there is no draft yet, stop: this skill edits, it does not originate.
- Distinctive: line 28
Read the draft and the platform.
- Distinctive: line 30
Run the anti-slop pass (see Guardrails).
- Distinctive: line 41
No em-dashes. No "in today's fast-paced world", "unlock", "delve", "tapestry", "game-changer", or filler transitions.
- Distinctive: line 43
Do not invent metrics, anecdotes, tools used, or feelings.
- Distinctive: line 44
Keep the owner's actual opinions.
- Very specific: line 48
After each run, add an entry to [[Skills/example-voice/FEEDBACK|FEEDBACK]].
≈ “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.
Files
Example Voice
Overview
A style layer that rewrites an existing draft into the owner's voice. It never generates the first draft's structure and never invents facts. Pair it with an output skill (post, script, newsletter).
Load First
<grounding skill>for the owner's background, opinions, and safe positioning.- The output skill that produced the draft (post / script / review).
Inputs
- A complete first draft to rewrite.
- The target platform (post, video, email) so rhythm and length fit.
- If there is no draft yet, stop: this skill edits, it does not originate.
Workflow
- Read the draft and the platform.
- Rewrite for voice: shorter sentences, concrete nouns, first person, no hype.
- Run the anti-slop pass (see Guardrails).
- Return the rewritten draft only. Show scores/notes only if asked.
Output
- The rewritten draft, saved in its project folder (see [[AGENT-RULES]] §1 and §3), not
working-dir/. - This skill produces notes, not code, so nothing gets pushed to a repo. (Code tasks keep their code in an external repo; see [[AGENT-RULES]] §7.)
Guardrails
- No em-dashes. No "in today's fast-paced world", "unlock", "delve", "tapestry", "game-changer", or filler transitions.
- Do not invent metrics, anecdotes, tools used, or feelings.
- Keep the owner's actual opinions; do not sand off the edges into neutral prose.
Feedback log
After each run, add an entry to [[Skills/example-voice/FEEDBACK|FEEDBACK]]. When a lesson repeats, promote it into this file and bump updated:. Rules: [[AGENT-RULES]] §5.
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.

