Anthropic-Cybersecurity-Skills copilot-instructions.md

A set of contribution instructions for the Anthropic-Cybersecurity-Skills repository, a library of cybersecurity procedures for AI agents. It explains the repository layout, required format, and validation steps.

In plain words
What is it for?
Adding or editing cybersecurity skills, understanding the repository structure, checking frontmatter, and running validation before submitting changes.
Why use it?
It gives contributors consistent rules for finding, writing, and checking skills, reducing errors when the library changes.

Instructions file for GitHub Copilot

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/mukul975/anthropic-cybersecurity-skills/copilot-instructions
Clone the repo
git clone --depth 1 https://github.com/mukul975/Anthropic-Cybersecurity-Skills

Made for: GitHub Copilot.

Per session 4,437 This file is loaded in full into every session.
When invoked 4,437 The same file — it is already loaded in full.
Security scan A 1 finding. 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 $0.04437 $0.04437
Opus 5 $0.02218 $0.02218
Sonnet 5 $0.00887 $0.00887
Haiku 4.5 $0.00444 $0.00444

Measured yesterday against content hash e17ff31d8586, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Anthropic-Cybersecurity-Skills copilot-instructions.md scanned grade A with 1 finding 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 yesterday.

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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Use standard Unix tools (bash, Python, curl, etc.)
.github/copilot-instructions.md · 547 lines

How it starts

The opening of the file, as written. The whole thing — 547 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Copilot Instructions for Anthropic-Cybersecurity-Skills

Anthropic-Cybersecurity-Skills is the largest open-source cybersecurity skills library for AI agents, containing 817 production-grade skills mapped to 6 industry frameworks (MITRE ATT&CK, NIST CSF, MITRE ATLAS, MITRE D3FEND, NIST AI RMF, MITRE F3). This document guides Copilot agents contributing new skills and framework mappings.

Quick Facts

  • Type: Open-source cybersecurity skills library + framework mapping hub
  • Skills: 817 across 29 security domains
  • Frameworks: 6 (MITRE ATT&CK v14, NIST CSF 2.0, MITRE ATLAS, MITRE D3FEND, NIST AI RMF, MITRE F3)
  • Format: agentskills.io standard (YAML frontmatter + Markdown)
  • License: Apache 2.0 (ethical use required)
  • Community: Independent, community-created (not affiliated with Anthropic)

Repository Structure

Anthropic-Cybersecurity-Skills/
├── skills/                    # 817 skill directories (kebab-case)
│   ├── abusing-dpapi-for-credential-access/
│   │   ├── SKILL.md          # Frontmatter + detailed instructions
│   │   ├── LICENSE
│   │   ├── scripts/
│   │   │   └── process.py    # Optional helper scripts
│   │   └── references/
│   │       ├── api-reference.md
│   │       ├── standards.md
│   │       └── workflows.md
│   └── ... (816 more)
├── mappings/                  # Framework coverage & alignment
│   ├── mitre-attack/
│   │   ├── attack-navigator-layer.json
│   │   └── coverage-summary.md
│   ├── nist-csf/
│   ├── owasp/
│   └── README.md
├── docs/                      # Additional documentation
├── index.json                 # Central skill registry (auto-generated)
├── CONTRIBUTING.md            # Contribution guide
├── SECURITY.md               # Ethical use & dual-use policies
└── CODE_OF_CONDUCT.md        # Community guidelines

Build & Development

Prerequisites

  • Git (for cloning and version control)
  • Python 3.8+ (optional, for scripts/metadata generation)
  • jq (optional, for JSON processing; useful for index.json queries)

Read the full file on GitHub · 547 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. yesterday First seen · 547 lines · 4,437 tokens per session scan A e17ff31d8586

Subscribe to this mod's changes

Anthropic-Cybersecurity-Skills copilot-instructions.md is an instructions file published in the GitHub repository mukul975/Anthropic-Cybersecurity-Skills (31,693 stars, last pushed 8d ago), licensed Apache-2.0. It adds 4,437 tokens to every session, about $0.0222 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.