The quality-governed, cross-runtime skill collection. Agent skills as first-class software assets — with evidence-based validation, community audit, and lifecycle governance.
monna-os-skills is a curated, open-source collection of AI agent skills that treats skills as auditable software assets — not disposable prompts.
Every skill in this collection passes a rigorous validation pipeline: structure checks, overlap detection, security scanning, content review, and community audit. Skills are organized by quality tier, not by download count.
The agent skill ecosystem is exploding — but quality is cratering. Most registries (52K+ skills on some platforms) are open dumps with zero gatekeeping. Documented malware exists. Conflicting skills bloat agent context. No one treats skills as governed, versioned, lifecycle-managed assets.
monna-os-skills fixes that by building the infrastructure layer the ecosystem desperately needs:
- Evidence-gated validation — every claim traceable, every overlap blocked
- Quality tiers — Draft → Validated → Reviewed → Certified → Universal
- Lifecycle governance — deprecation, migration, sunset paths
- Cross-runtime compatibility — tested on Kimi/Daimon, Claude, Copilot, Cursor, OpenClaw, Gemini
# Using Vercel skills CLI (universal)
npx skills add emanalshazly/monna-os-skills
# Using GitHub CLI (GitHub Copilot)
gh skill add emanalshazly/monna-os-skills
# Or manually copy the SKILL.md file to your agent's skills directory| Domain | Skills | Quality Tier |
|---|---|---|
| AI Governance & Compliance | 3 | 🏆 Certified |
| Security & Risk Assessment | 2 | 🔍 Reviewed |
| Data Governance | 1 | ✅ Validated |
| Infrastructure & DePIN | 1 | 🏗️ Draft |
| Multilingual & Cross-Cultural | 1 | ✅ Validated |
| Open-Source Tooling | 2 | 🏆 Certified |
| Tier | Badge | What It Means |
|---|---|---|
| 🏗️ Draft | draft |
Submitted, under review. Not yet validated. |
| ✅ Validated | validated |
Passes automated structure, overlap, and security checks. |
| 🔍 Reviewed | reviewed |
Community audit completed. No structural or security issues. |
| 🏆 Certified | certified |
Multiple independent audits. Real-world usage confirmed. |
| 🌐 Universal | universal |
Certified AND tested on 5+ runtimes. |
All skills follow the Anthropic Agent Skills spec with MONNA extensions for quality governance:
---
name: example-skill
version: 1.0.0
description: "Use when..."
categories: [governance, compliance]
quality_tier: certified
compatibility:
kimi: certified
claude: certified
copilot: validated
fingerprint: "gov-001-a7f3c9"
last_tested: 2026-06-24
---See Skill Authoring Guide for full details.
We welcome contributions! Please read our Contributing Guide and Governance Model before submitting.
Key rule: All skills must be distinct from the commercial monna-skill-portfolio (sold on PromptBase/Gumroad). This repo is permanently free and open-source.
Found a security issue in a skill? See our Security Policy. We take security seriously — skills execute with agent privileges, so we audit them like code.
MIT — permanently free, no attribution required for usage.
Built and maintained by monna (@emanalshazly) — prompt engineer, AI governance practitioner, and open-source contributor.
"Power without governance is chaos. Let's build the governance layer together."