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/shyuan/writing-humanizer/agents-mdgit clone --depth 1 https://github.com/shyuan/writing-humanizerWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/shyuan/writing-humanizer/agents-md)<a href="https://agentmods.dev/instructions/shyuan/writing-humanizer/agents-md"><img src="https://agentmods.dev/badge/instructions/shyuan/writing-humanizer/agents-md.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.01327 | $0.01327 |
| Opus 5 | $0.00664 | $0.00664 |
| Sonnet 5 | $0.00265 | $0.00265 |
| Haiku 4.5 | $0.00133 | $0.00133 |
Grade A, and why
writing-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 5d 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
What this repository is
A cross-tool plugin (not an application). It ships a single user-invocable skill, writing-humanizer, that detects and rewrites AI-generated writing patterns, focused on Traditional Chinese (Taiwan). There is no build, lint, test, or runtime — the "code" is Markdown prompt content that the host (Claude Code, OpenAI Codex, Cursor, Google Antigravity CLI, OpenCode, Kimi CLI, or pi) loads as a skill. Changes are validated by reading them and by exercising the skill, not by a toolchain.
The same skills/ directory and SKILL.md format are shared by all hosts (Claude Code, OpenAI Codex, Cursor, Google Antigravity CLI, OpenCode, Kimi CLI, pi); only the manifest/loader differs per host.
Repository layout
.claude-plugin/plugin.json— Claude Code plugin manifest. Distributed via theshyuan-marketplace..codex-plugin/plugin.json— OpenAI Codex plugin manifest. Points at the shared skills dir via"skills": "./skills/"..cursor-plugin/plugin.json— Cursor plugin manifest (Cursor 2.5+). Points at the shared skills dir via"skills": "skills/".plugin.json(repo root) — Google Antigravity CLI plugin manifest. Antigravity expects this at the plugin root (not in a namespaced dir);"skills"is an array pointing at each skill dir (["skills/writing-humanizer"]).package.json+.opencode/plugins/writing-humanizer.js+.opencode/INSTALL.md— OpenCode loader. OpenCode does not read the top-levelskills/dir; instead it installs this repo as a JS plugin (via thepluginarray in the user'sopencode.json, resolved throughpackage.json's"main"), and the plugin'sconfighook injectsskills/intoconfig.skills.pathsso the skill is auto-discovered — no symlinks. See.opencode/INSTALL.md..kimi-plugin/plugin.json— Kimi CLI plugin manifest. Points at the shared skills dir via"skills": "./skills/", plus a Kimi-specificinterfaceblock for display metadata. (We omit Kimi's optionalsessionStartso the skill loads on demand, andskillInstructionssince the skill references no host-specific tools.).pi/extensions/writing-humanizer.ts+ the"pi"field inpackage.json— pi loader. Like OpenCode, pi does not read the top-levelskills/dir on its own; the TS extension'sresources_discoverhook returnsskillPaths: [skillsDir]to register the sharedskills/dir.package.json's"pi"field lists the extension and skills paths.skills/writing-humanizer/SKILL.md— the skill entry point (hub), used by all seven hosts. Its YAML frontmatter (name,description) drives triggering on all of them; Claude-specific keys (user-invocable,argument-hint,allowed-tools) are ignored by the others.skills/writing-humanizer/references/— spoke files loaded on demand by the hub.README.md— bilingual (English + 中文) user-facing docs, kept in sync with the pattern catalog.
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.
- 5d ago First seen · 47 lines · 1,327 tokens per session scan A 92d2d6161b53
writing-humanizer AGENTS.md is an instructions file published in the GitHub repository shyuan/writing-humanizer (26 stars, last pushed 2mo ago), licensed MIT. It adds 1,327 tokens to every session, about $0.0066 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 2026-08-30.
Other instructions, from other repositories
shuorenhua AGENTS.md
AGENTS.md instructions for MrGeDiao/shuorenhua, covering 仓库说明, 语言, 风格, 布局 and 协作.
shuorenhua CLAUDE.md
Claude Code instructions for MrGeDiao/shuorenhua, a project described as: 说人话|中文优先的去 AI 味改写 skill:保事实、分场景、改完可直接发。Chinese-first rewrite skill for Codex / Claude Code / Cursor / ChatGPT — removes AI tone, preserves facts.
superpowers-zh CLAUDE.md
Claude Code instructions for jnMetaCode/superpowers-zh, covering superpowers — 贡献者指南, 如果你是 ai agent, pull request 要求, 我们不会接受的内容 and 第三方依赖.
avoid-ai-writing CLAUDE.md
Claude Code instructions for conorbronsdon/avoid-ai-writing, covering claude.md, what this is, repository structure, how to make changes and architecture of the skill.
spec-kit-zh AGENTS.md
AGENTS.md instructions for loulanyue/spec-kit-zh, covering agents.md, about spec kit and specify, general practices, adding new agent support and current supported agents.
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.