Open Agent Skills by CIRSEC: hardening, security operations and threat intel workflows.
Each skill is a folder under skills/ with a SKILL.md (YAML frontmatter + instructions). The format is an open standard, so these skills work with Claude (Code, desktop, claude.ai), Codex, Cursor, Gemini CLI, Copilot and any agent that reads SKILL.md.
| Skill | What it does |
|---|---|
cir |
Read, write, validate, score and share CIR security issue records, and build software compatible with the format. Ships the v0.5 spec, schema, examples and a reference helper (scripts/cir.py). |
cirsec-baselines |
Understand and use CIRSEC Baselines: hardening recommendations, security configuration assessments with findings and fixes, framework gap analysis (NIST, ANSSI, CISA, ATT&CK), tailoring, scan results, writing controls and building compatible scanners. Uses the public MCP server, with a stdlib fallback client (scripts/baselines.py). |
/plugin marketplace add cirsec/skills
/plugin install cirsec-skills@cirsec
Update later with /plugin marketplace update cirsec.
The plugin also connects the two public, read-only MCP servers the skills rely on: https://cirsec.org/mcp (CIR) and https://baselines.cirsec.org/mcp (Baselines). With another agent, add them as remote MCP servers yourself.
npx skills add cirsec/skills --list
npx skills add cirsec/skills --skill '*'
Download a skill zip from the latest release and upload it in the Skills settings.
cp -r skills/<skill-name> ~/.claude/skills/ # global
cp -r skills/<skill-name> .claude/skills/ # per project
See CONTRIBUTING.md.
Apache-2.0