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 skills add Ghosteken/agent-harness --skill ai-mdgit clone --depth 1 https://github.com/Ghosteken/agent-harnessWrote 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/skills/ghosteken/agent-harness/ai-md)<a href="https://agentmods.dev/skills/ghosteken/agent-harness/ai-md"><img src="https://agentmods.dev/badge/skills/ghosteken/agent-harness/ai-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 5.1 | $0.00033 | $0.04987 |
| Opus 5 | $0.00016 | $0.02493 |
| Sonnet 5 | $0.00007 | $0.00997 |
| Haiku 4.5 | $0.00003 | $0.00499 |
Grade B, and why
ai-md scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
rules=$(cat ~/.claude/rules/*.md 2>/dev/null | wc -c || echo 0) Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
EVIDENCE: no-fabricate no-guess | 禁用詞:應該是/可能是 → 先拿數據 | Read/Grep→行號 curl→數據 | "好像"/"覺得"→自己先跑test | guess=shame-wall This is a copy
100% identical to ai-md — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 524 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI.MD v4 — The Complete AI-Native Conversion System
When to Use This Skill
- Use when your CLAUDE.md is long but AI still ignores your rules
- Use when token usage is too high from verbose system instructions
- Use when you want to optimize any LLM system prompt for compliance
- Use when migrating rules between AI tools (Claude, Codex, Gemini, Grok)
What Is AI.MD?
AI.MD is a methodology for converting human-written CLAUDE.md (or any LLM system instructions)
into a structured-label format that AI models follow more reliably, using fewer tokens.
The paradox we proved: Adding more rules in natural language DECREASES compliance. Converting the same rules to structured format RESTORES and EXCEEDS it.
Human prose (6 rules, 1 line) → AI follows 4 of them
Structured labels (6 rules, 6 lines) → AI follows all 6
Same content. Different format. Different results.
Why It Works: How LLMs Actually Process Instructions
LLMs don't "read" — they attend. Understanding this changes everything.
Mechanism 1: Attention Splitting
When multiple rules share one line, the model's attention distributes across all tokens equally. Each rule gets a fraction of the attention weight. Some rules get lost.
When each rule has its own line, the model processes it as a distinct unit. Full attention weight on each rule.
# ONE LINE = attention splits 5 ways (some rules drop to near-zero weight)
EVIDENCE: no-fabricate no-guess | 禁用詞:應該是/可能是 → 先拿數據 | Read/Grep→行號 curl→數據 | "好像"/"覺得"→自己先跑test | guess=shame-wall
# FIVE LINES = each rule gets full attention
EVIDENCE:
core: no-fabricate | no-guess | unsure=say-so
banned: 應該是/可能是/感覺是/推測 → 先拿數據
proof: all-claims-need(data/line#/source) | Read/Grep→行號 | curl→數據
hear-doubt: "好像"/"覺得" → self-test(curl/benchmark) → 禁反問user
violation: guess → shame-wall
Mechanism 2: Zero-Inference Labels
Natural language forces the model to INFER meaning from context. Labels DECLARE meaning explicitly. No inference needed = no misinterpretation.
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.
- 7d ago First seen · 524 lines · 33 tokens per session scan B b322fcd60bf7
ai-md is a skill published in the GitHub repository Ghosteken/agent-harness (2 stars, last pushed 3d ago), licensed MIT. It adds 33 tokens to every session and 4,987 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). It is 100% identical to ai-md, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
agent-self-scheduling
Schedule AI agent runs with cron, loops, or external clocks while avoiding unsafe tight autonomous timers.
serial
A tool for communicating with devices over serial ports, such as USB-to-UART connections. It can find ports, watch incoming data, send messages, save logs, and display raw bytes in hexadecimal form.
eide
A build tool for EIDE projects, an embedded-development extension for Visual Studio Code. It finds EIDE project settings and builds firmware using ARM CC or GCC.
net
A toolkit for diagnosing network connections and traffic in embedded systems. A packet capture is a recorded sample of network traffic, and this tool can create or analyze captures in formats such as pcap and pcapng.
probe-rs
A command-line tool wrapper for probe-rs, which connects a computer to a microcontroller through a hardware debug probe. It supports common firmware programming and debugging operations over SWD or JTAG.
terminal
An interactive terminal session that keeps a live connection to a local shell, an SSH remote computer, or a serial device. It lets you send commands and read the next output while preserving the session’s login, menu, or program state.