evaluate-security-risk

Security Risk Review

Produces a STRIDE threat model, flagging the risks teams commonly miss.

Part of the expert-intelligence-skills marketplace.

What you get

A threat model over your system — STRIDE by default, or another framework if you name one — with each threat tied to published analysis rather than a generic checklist, and explicit attention to the categories teams routinely skip.

Output is grounded in what practitioners have actually published — every claim carries a citation back to the O'Reilly learning platform, and the skill is instructed never to cite a title, author, or link that didn't surface in its search.

Install

Expand the path that matches your setup. Whichever you use, restart your session or app afterward so the new skill is picked up.

Claude desktop

1. Open Customize, then Plugins.
2. Click +, then Add marketplace, then Add from a repository.
3. Paste oreillymedia/expert-intelligence-skills.
4. Install Security Risk Review from the list.

Anthropic's walkthrough has the current UI: https://support.claude.com/en/articles/13837440-use-plugins-in-claude

Claude CLI

From inside a session:

/plugin marketplace add oreillymedia/expert-intelligence-skills
/plugin install evaluate-security-risk@expert-intelligence-skills

Codex

For both the Codex CLI and Codex desktop, you must use the terminal to add the marketplace — run:

codex plugin marketplace add oreillymedia/expert-intelligence-skills
codex plugin add evaluate-security-risk@expert-intelligence-skills

Manual: copy the file — works anywhere

Download SKILL.md with the Download raw file button, then:

• Claude desktop, web, or Cowork: open Customize → Skills, click Add → Upload skill, and drop in the .md file — it's accepted as-is, no zipping needed. One upload covers both Chat and Cowork.
• Claude Code or Codex CLI: save it, filename unchanged, as evaluate-security-risk/SKILL.md inside ~/.claude/skills/ or ~/.codex/skills/. This route is CLI-only — the desktop and web apps don't read those directories.

Important: When using this method you won't receive updates to this skill.

Requirements

An O'Reilly learning platform account is required to search the library. When installed as a plugin, the O'Reilly Expert Intelligence MCP server is included automatically. If you copy an individual skill manually instead, connect the MCP server yourself; see the MCP guide for setup steps.

Try it

Threat model our new payments endpoint.

What security risks are we missing here?

STRIDE review of this auth flow.

This skill triggers based on the shape of your context and prompt. See the description field in skills/evaluate-security-risk/SKILL.md for the full set of trigger conditions.