ai-act-skill AGENTS.md

ai-act-skill AGENTS.md is an instructions file for Codex, OpenCode from morellid/ai-act-skill. It costs 1,830 tokens per session, scanned A, original, MIT.

An instruction set for coding agents that helps with compliance questions about the European Union’s AI Act, a law governing artificial-intelligence systems. It covers how to classify systems and check different legal obligations.

In plain words
What is it for?
Use it to classify an AI system, check prohibited practices, and review duties for high-risk systems, deployers, and providers of general-purpose AI models.
Why use it?
It gives an agent a defined way to handle AI Act questions instead of treating the repository like application code. This helps organize compliance work around the relevant parts of the law.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/morellid/ai-act-skill/agents-md
Clone the repo
git clone --depth 1 https://github.com/morellid/ai-act-skill

Made for: Codex, OpenCode.

Wrote 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.

agentmods badge for ai-act-skill AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/morellid/ai-act-skill/agents-md.svg)](https://agentmods.dev/instructions/morellid/ai-act-skill/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/morellid/ai-act-skill/agents-md"><img src="https://agentmods.dev/badge/instructions/morellid/ai-act-skill/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,830 This file is loaded in full into every session.
When invoked 1,830 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.01830 $0.01830
Opus 5 $0.00915 $0.00915
Sonnet 5 $0.00366 $0.00366
Haiku 4.5 $0.00183 $0.00183

Measured 5d ago against content hash 4fc743d4d10a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ai-act-skill 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.

AGENTS.md · 111 lines

How it starts

The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md — AI Act Compliance Skill

Open-format guidance for AI coding agents. AGENTS.md is supported natively by Codex, Cursor, Windsurf, GitHub Copilot, Devin, Amp, Antigravity, and others. See https://agents.md.

What this repository is

This repository is an agent skill for compliance work on Regulation (EU) 2024/1689 — the EU AI Act. Its purpose is to be installed as a skill into a coding/agent runtime, then invoked when the user asks AI Act questions.

It is not application code. There are no tests to run, no build to perform, no production deployment. Working on this repository means improving the skill itself: the article extracts, the task instructions, the example fixtures.

How to use this skill from your agent

The skill is structured around six tasks routed by SKILL.md:

  • tasks/classify-system.md — classify an AI system / model under the AI Act
  • tasks/check-prohibited-practices.md — Article 5 (eight prohibited practices)
  • tasks/check-high-risk-provider.md — Articles 8–22
  • tasks/check-deployer-obligations.md — Articles 26–27 (incl. FRIA)
  • tasks/check-gpai-provider.md — Articles 51–55 (incl. systemic risk threshold)
  • tasks/check-transparency.md — Article 50

The textual extracts of cited articles live in references/extracts/. The references/sources.yaml file tracks each source with its URL, accessed date, SHA256 hash, and licence.

If you are running this repo as a skill (Claude Code, Codex)

Install it into your skills directory and let your agent discover it natively:

# Claude Code
git clone https://github.com/morellid/ai-act-skill.git ~/.claude/skills/ai-act-compliance

# Codex
git clone https://github.com/morellid/ai-act-skill.git ~/.agents/skills/ai-act-compliance

See README.md for the full installation matrix.

If you are working on this repository (improving the skill)

Conventions:

  • Source-grounded: every normative claim must trace to the Regulation, an EU Commission delegated/implementing act, an EU AI Office guideline, an EDPB joint guideline, or a published harmonised standard. Cite the article precisely.
  • No fabrication: if a fact is not in references/extracts/ or references/sources.yaml, do not invent it. Add the source first, then cite it.
  • Disclaimer always present: every task file and every output template includes the language "this is a support tool, not legal advice" and points the user to qualified counsel.
  • Plain markdown: tasks and extracts are plain markdown. No HTML, no images, no executable inline content. Code blocks are illustrative output templates, not runnable code.
  • Italian: not in this repo. The Italian-market variant of the skill lives separately. This repo is English, EU-wide audience.

Read the full file on GitHub · 111 lines

Changes

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.

  1. 5d ago First seen · 111 lines · 1,830 tokens per session scan A 4fc743d4d10a

Subscribe to this mod's changes

ai-act-skill AGENTS.md is an instructions file published in the GitHub repository morellid/ai-act-skill (4 stars, last pushed 27d ago), licensed MIT. It adds 1,830 tokens to every session, about $0.0092 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.

Related

Other instructions, from other repositories