Skills
Packaged AI-agent skills built on xgrep — code understanding, rule authoring, triage, fixing, and secure coding, for Claude Code or Codex.
A set of AI-agent skills built on xgrep is published in the
mondoohq/secure-stack marketplace. Each is a
self-contained plugin with a README.md, a SKILL.md, command definitions, and
reference material, installable in Claude Code, Codex, Gemini CLI, or Cursor.
| Skill | What it does |
|---|---|
| understand-code | Navigate and understand source code using xgrep's AST-powered code intelligence. |
| author-detections | Create custom xgrep rules with a test-first methodology, including taint rules, or port rules to new languages. |
| triage-findings | Investigate and classify scan findings using code-graph analysis. |
| fix-findings | Fix one confirmed finding, or a whole set of them, via the verify/apply harness. |
| secure-coding | Proactively avoid generating vulnerable code. |
| secure-pipeline | Scan, triage and fix before a change ships. |
Installing the skills
Claude Code
Register the marketplace once, then install the skills you want:
/plugin marketplace add mondoohq/secure-stack
/plugin install triage-findings@secure-stack
/plugin install author-detections@secure-stackOnce installed, invoke a skill directly (/triage-findings, /author-detections, …) or let it
activate automatically on a relevant task.
Codex, Gemini CLI, Cursor
The same skills install into Codex, Gemini CLI, and Cursor. See the mondoohq/secure-stack README for the per-agent instructions.
For the underlying CLI these skills drive, see Code intelligence and the AI agent guide.
Last updated on