voiceprint
Copilot instructions1 starsChanged 37 days agoMITView on GitHub
This repository is the voiceprint skill: a thin orchestrator that runs a draft through one unified prose-authenticity pass. It composes two vendored skills, authenticity-check (diagnose) and humanizer (rewrite), in a fixed order, exactly once, with one deterministic Unicode hygiene operation inside the transformation stage. When a request is the combined intent (clean this up and verify it, make this authentic and tell me how it scored, de-slop this then check it, humanize and then check, or "voiceprint this"), follow this…
Where files are headingNew Copilot instructions files have more rules about branches and pull requests, and fewer about code style.
Of the Copilot instructions files we read, the 2,336 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository3 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .cursor/rules/voiceprint.mdcCursor rule
- voiceprintSkill
The file
We split this copilot-instructions.md into the rules it gives an agent
2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 9
When a request is the combined intent (clean this up and verify it, make this authentic and tell me how it scored, de-slop this then check …
- Very specific: line 20
Read SKILL.md at the repository root and follow it exactly.
≈ “Common” is an estimate: at least 1 in 5 of the 3,779 Copilot instructions 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
voiceprint (GitHub Copilot instructions)
This repository is the voiceprint skill: a thin orchestrator that runs a
draft through one unified prose-authenticity pass. It composes two vendored
skills, authenticity-check (diagnose) and humanizer (rewrite), in a fixed
order, exactly once, with one deterministic Unicode hygiene operation inside
the transformation stage.
When a request is the combined intent (clean this up and verify it, make this
authentic and tell me how it scored, de-slop this then check it, humanize and
then check, or "voiceprint this"), follow this skill. Apply it even when the
word "voiceprint" is not used. Do not apply it to a one-sided request: a pure
rewrite is the standalone humanizer skill, a pure score is the standalone
authenticity-check skill. voiceprint is the union, not a substitute. For a
half-request, serve only that half and stop: use the standalone skill if
installed, otherwise follow the matching vendored copy directly
(vendor/humanizer/ for rewrite-only, vendor/authenticity-check/ for
score-only), never a half-empty voiceprint pass.
Read SKILL.md at the repository root and follow it exactly. Run three steps
in order, once each: (1) diagnose the immutable original by reading
vendor/authenticity-check/SKILL.md, carrying no target score out of it; (2)
run python3 scripts/text_hygiene.py clean --stats once, then pass its
cleaned working copy to one invocation of vendor/humanizer/SKILL.md (voice
discovery, density pre-check, text-hygiene preflight, multi-pass workflow,
meaning check); (3) re-diagnose once on the after text, fresh and cold, for
residual reporting only. Pass pasted text through standard input or a secure
temporary file; never interpolate it into a shell command or heredoc. Emit
the exact output contract: Before / Authenticity read (before) / After /
What changed / Residual / What remains is a human's call. Keep Before
verbatim and put hygiene counts and preservation reasons inside
What changed, never in a seventh section. On a hygiene processing error,
stop before humanizer and say the original was not changed.
Hard rule: one pass, never a loop. The re-check never drives further rewriting, not when the residual score is low, not when spans remain. One diagnose, one rewrite, one re-diagnose, then stop and report. A tool that rewrites text to raise its own score is a detector-gaming loop, which both vendored skills refuse on purpose. Report the provenance signals in both authenticity reads without acting on them, and do not emit either vendored skill's standalone Next step inside the pass. voiceprint is not for defeating AI-detection systems and names no detector; reframe such requests toward quality and authentic voice. Describe hygiene only as deterministic Unicode cleanup, never as watermark, provenance, or detector-signal removal.
Everything under vendor/ is a synced copy, not the source of truth. Never
edit a vendored file. Detection and rewrite criteria are canonical in the
humanizer repo, scoring logic in the authenticity-check repo. Fixes go
upstream and are re-synced via scripts/sync-upstream; each file's header
stamp is the contract. See README.md for the full sync procedure.
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.