Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/hannsxpeter/humanizer/agents-mdgit clone --depth 1 https://github.com/hannsxpeter/humanizerWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00676 | $0.00676 |
| Opus 5 | $0.00338 | $0.00338 |
| Sonnet 5 | $0.00135 | $0.00135 |
| Haiku 4.5 | $0.00068 | $0.00068 |
Grade A, and why
humanizer AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
humanizer (agent instructions)
This repository is the humanizer skill: a pure-prompt instruction set that
rewrites AI-sounding prose so it reads as genuinely human, and rewrites in a
specific writer's voice when a sample or profile is available. It also
conservatively cleans suspicious invisible Unicode in supplied prose. No
scripts, no dependencies, no network access. It is the entry point for any AI
coding tool that reads AGENTS.md (Codex, OpenCode, Antigravity, Pi Coder, and
others).
When to apply this skill
Apply it whenever the user wants to humanize, de-slop, de-AI, or de-robotify text; to fix writing that sounds like an LLM, corporate, salesy, generic, or "off"; to make a draft sound like them or like a named author; or to edit prose for authentic voice and rhythm. Apply it even when they do not say the word "humanize" and do not name this skill. Apply its text-hygiene pass when prose may contain zero-width characters, unusual spaces, direction controls, or copy-paste residue.
How to run it
Read SKILL.md in this repository and follow it exactly. The full method
lives there and in references/; do not improvise a shortcut. In brief:
- Step 0 discover the voice (pasted sample, named author, or a
VOICE.md/STYLE-GUIDE.mdin the working tree). No voice found means generic mode; never invent a persona. - Step 0b stance mode is off by default. Turn it on only if the user explicitly asks for more voice, edge, or opinion.
- Step 0c density pre-check: skim for dead-giveaway tells and pick a light, standard, or full pass so human-first text is not over-edited.
- Step 0d text hygiene: inspect conservatively, remove only characters with no semantic role, and preserve load-bearing Unicode.
- Multi-pass: voice injection (if a voice exists), then tell removal
against
references/tell-patterns.md(32 patterns, six families), then a self-audit againstreferences/do-not-flag.md. - Emit the exact output contract from
SKILL.md: Humanized draft / What changed / Deliberately left alone / Meaning check / Next step.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 55 lines · 676 tokens per session scan A 47edf964fac8
humanizer AGENTS.md is an instructions file published in the GitHub repository hannsxpeter/humanizer (7 stars, last pushed 17d ago), licensed MIT. It adds 676 tokens to every session, about $0.0034 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
authenticity-check AGENTS.md
Instructions for hannsxpeter/authenticity-check, covering authenticity-check (agent instructions), when to apply this skill, how to run it, hard rule (diagnostic only) and scope.
authenticity-check copilot-instructions.md
Instructions for hannsxpeter/authenticity-check: This repository is the authenticity-check skill: pure-prompt instructions that score how authentically a piece of text reads as the work of a real human author and flag the spans that read as AI-generated, AI-templated, or generically derivative, plus suspicious Unicode…
authenticity-check GEMINI.md
Instructions for hannsxpeter/authenticity-check: This repository is the authenticity-check skill: pure-prompt instructions that score how authentically a piece of text reads as the work of a real human author and flag the spans that read as AI-generated, AI-templated, or generically derivative, plus suspicious Unicode…
vale-ai-tells CLAUDE.md
Instructions for tbhb/vale-ai-tells, covering claude code instructions, project overview, repository structure, development workflow and rule conventions.
vale-ai-tells AGENTS.md
Instructions for tbhb/vale-ai-tells, covering agent instructions, commit messages, prose lint output and drafting a document.
humanizer-skill AGENTS.md
AGENTS.md instructions for Aboudjem/humanizer-skill, covering agents.md — humanizer skill, what this repo is, file map, how to install the skill and how to invoke the skill.