llm-tells SKILL.md skillA

llm-tells from rikvanriel/code-hygiene is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add rikvanriel/code-hygiene/llm-tells@v1.1.0

Pins v1.1.0 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
rikvanriel/code-hygiene
Path
skills/llm-tells/SKILL.md
Versions recorded
5
Latest version
v1.1.0, committed Sep 17, 2026
First version recorded
Sep 16, 2026
Size
4,809 bytes
Repository stars
2
License
MIT

Use as a final pass before commit or PR, or when text sounds AI-generated. Strips AI tells from code, comments, changelogs.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# LLM Tells — Final Pass

Final-pass checklist for generic repo (derived from kernel-readability but project-agnostic). Run after self-review-gate Gate1 before finishing.

Scope: text your patch adds/changes. Pre-existing prose out of scope — rewriting it inflates diff and collides with later patches touching same lines. Fix in own commit.

## Verification — never invent

- [ ] Every number/quote/date/perf/hash/claim sourced this session (file, git log/show, cmd output, benchmark, crash dump, public tracker).
- [ ] If don't know, TODO — not plausible fill.
- [ ] Hashes exist in git log (`git show <hash>`) when trailer claims `Fixes: <hash>`.
- [ ] Links public and resolve — no private scheme (`file://`, local report path).
- [ ] Scope matches diff stat — no claims about untouched files.

## Changelog

- [ ] Opens with "This patch …" or fix → rewrite to open with problem/current behavior in present tense.
- [ ] Bugfix doesn't lead with real-world symptom (who hits, what breaks / how observed) → add it; race: CPU0/CPU1 ladder or sequence.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add rikvanriel/code-hygiene/llm-tells@sha256:be2d03bf104aba43

The line at the top of this page pins the label v1.1.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_7a2etbvpwwwiz2hb.svg)](https://markdownregistry.com/a/art_7a2etbvpwwwiz2hb)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 5 versions of this file, between Sep 16, 2026 and Sep 17, 2026, more than one a day.

versioncommittedcommitsizeaudit
v1.1.0 latest2026-09-17 f466313 4,809 BA view · diff
v1.1.02026-09-17 4fc5ed9 4,806 BA view · diff
v1.1.02026-09-17 6de94e8 4,770 BA view · diff
v1.1.02026-09-16 a69317f 4,800 BA view · diff
v1.0.02026-09-16 c53b7e1 4,196 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (4809 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

rikvanriel/code-hygiene · 2 stars · license MIT · pushed 2026-10-16 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_7a2etbvpwwwiz2hb
GET https://markdownregistry.com/api/v1/resolve?ref=rikvanriel/code-hygiene/llm-tells
GET https://markdownregistry.com/api/v1/blob/be2d03bf104aba4379cf05e211aa81048b49ff667b53ac6ea006cefadca0bc85

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from rikvanriel/code-hygiene

AGENTS.md AGENTS.md file
rikvanriel/code-hygiene · AGENTS.md
git:20260927.dfb67f5 · audit A · 2 stars
CLAUDE.md CLAUDE.md file
rikvanriel/code-hygiene · CLAUDE.md
git:20260916.7539a99 · audit A · 2 stars
artifact-handoff SKILL.md skill
rikvanriel/code-hygiene · skills/artifact-handoff/SKILL.md · Use when handing an artifact to someone, or trusting one you received — identity, provenance, verification.
v1.0.0 · audit A · 2 stars
catalog SKILL.md skill
rikvanriel/code-hygiene · skills/catalog/SKILL.md · Use when starting work in an unfamiliar repo, setting up hygiene skills, or unsure which skills a task needs…
v1.1.0 · audit A · 2 stars
change-splitting SKILL.md skill
rikvanriel/code-hygiene · skills/change-splitting/SKILL.md · Use when committing or splitting a large change — tests whether it splits into easy-to-review smaller changes: mixed…
v1.0.3 · audit A · 2 stars
changelog-quality SKILL.md skill
rikvanriel/code-hygiene · skills/changelog-quality/SKILL.md · Use when drafting or reviewing a commit message, changelog entry, diff summary, test plan, cover letter, or PR…
v1.0.1 · audit A · 2 stars
code-structure SKILL.md skill
rikvanriel/code-hygiene · skills/code-structure/SKILL.md · Use when a function is too long, nests too deep, or mixes themes. Extracts helpers that earn their names.
v1.0.1 · audit A · 2 stars
comment-quality SKILL.md skill
rikvanriel/code-hygiene · skills/comment-quality/SKILL.md · Use when writing or reviewing comments and docstrings — new, stale, or lacking WHY.
v1.0.2 · audit A · 2 stars
evidence-discipline SKILL.md skill
rikvanriel/code-hygiene · skills/evidence-discipline/SKILL.md · Use when measuring, benchmarking, or validating anything — proves the instrument, the sample, and the artifact.
v1.0.0 · audit A · 2 stars
factual-integrity SKILL.md skill
rikvanriel/code-hygiene · skills/factual-integrity/SKILL.md · Use when writing or verifying claims — numbers, hashes, quotes, perf results. Sources or TODOs it.
v1.0.0 · audit A · 2 stars
language-style-sampling SKILL.md skill
rikvanriel/code-hygiene · skills/language-style-sampling/SKILL.md · Use when working in an unfamiliar language or matching an unfamiliar codebase style. Derives a style guide by sampling.
v1.0.0 · audit A · 2 stars
plan-iteration-gate SKILL.md skill
rikvanriel/code-hygiene · skills/plan-iteration-gate/SKILL.md · Use when refining a draft plan, stuck iterating, or deciding whether to escalate to a human. Converges via 4Q loop.
v1.2.0 · audit A · 2 stars

Every file in rikvanriel/code-hygiene

Browse by kind, by grade A, or by owner.