THIS IS F**KING INSANE Found something genuinely comprehensive: ANTHROPIC CYBERSECURITY SKILLS 817 production grade security skills for AI agents, open source. 29 security domains —>cloud security—>threat hunting—>network security—>incident response—>malware analysis, red teaming, and more all mapped against real frameworks: MITRE ATT&CK v17.1, NIST CSF 2.0, OWASP, MITRE Fight Fraud... Framework, CIS Controls, and more. Each skill ships with full YAML frontmatter, references the exact framework techniques it covers, and works across Claude Code, GitHub Copilot, OpenAI Codex CLI, Cursor, and Gemini CLI. Quick start is a one line clone. Free, open source, actively maintained. Repo belowshow more

Repo:

Open source is running circles around proprietary security SaaS right now

Facts. The repo alone proves it.

you always find something interesting

Will keep looking for good stuff 🫡

Great resource for security workflows — and worth one clarification: it’s community-built, not an official Anthropic project. A one-line install also expands your agent’s trust boundary. Review the instructions, scripts, and dependencies. Pin the commit. Enable only what you need, with least privilege.

Absolutely. One line installs can hide a lot pin commits, review dependencies, and keep permissions minimal.

817 production-grade security skills in one repo might be the best free upgrade you can give an AI agent rn

Totally which security domain are you planning to test first with these skills?

this looks solid share bro

Thanks Roan

ty legend thanks for sharing

You welcome 🤗

Pretty insane

This is actually wild 817 production-ready security skills mapped to MITRE, NIST, OWASP and more, free and open source? Instant clone material

Go for it champ 🫡

it probably works really well

Have you tried it out

not yet

bookmarked, works with claude code + copilot + codex + cursor + gemini CLI is the whole stack

Yeah, that’s basically the whole stack

Safety skills are very important these days; thank you for sharing

Glad you find it useful 🤝

That's crazy fr

Is there a tool to check this repository for malicious or "hidden" instructions?

damn

Anthropic open-sourced 817 security skills. the real story is they stopped charging for what google charges $200/mo for. open source just ate the moat

great one man this is a wide take

Appreciate it, glad you liked it

real gold right there

@_0xpainn Happy you like it champ 🫡

My skill repo:

This is real f**king gold

Glad you like it champ

817 skills is a lot of coverage. The gap I keep seeing is the identity the agent assumes, not the playbook it runs. If the skill can open cloud and SaaS, who owns that non-human identity when the run ends?

Well said champ

817 skills mapped to ATT&CK is a solid blueprint. The gap is runtime enforcement. Without detection logic that fires when an agent deviates from those skills in production, you have a compliance doc, not a security control.

Eight hundred seventeen cybersecurity skills can be a remarkable reference library, but quantity is not the control. The useful test is whether an agent can select the right skill under ambiguity, refuse unsafe scope, and leave a reviewable trace. A catalog without enforcement can produce very confident coverage theater.

Production grade cybersecurity skills for agents, framework mapped and free.

@gork 这是什么项目

Nice share max

Damn

That single line install and the framework mappings make this super handy.

That’s bever gonna beat a real cybersecurity library like and its sibling adapters

The framework mappings make this far more usable than a pile of disconnected security prompts.

framework mapping is honestly the best part.

This is actually insane bro Thanks for sharing

@0x_sakata You welcome 🤗 buddy

A real find, thank you

Thanks for sharing max

817 skills sounds wild. Which security domain has held up best against a real repo so far, not just a clean test case?