AI agents hallucinate. They generate code without requirements, skip testing, make silent assumptions about hardware, and deploy to production without approval. Great for demos. Catastrophic for real products.
The Agile V™ Framework transforms unreliable AI agents into Verifiable Engineering Systems through:
- ✅ Formal Traceability — Every line of code links to
REQ-XXXX→ART-XXXX→TC-XXXX - ✅ Independent Verification — Red Team Verifier tests what Build Agent creates (no self-grading)
- ✅ Hardware Awareness — Agents validate RAM/CPU/GPU constraints before optimizing
- ✅ Human Gates — Evidence Summaries before production deployments (no autonomous releases)
- ✅ Halt on Ambiguity — Agents stop and ask when requirements are unclear
- ✅ Compliance-Ready — Auto-generates ISO 9001, ISO 27001, GxP artifacts from day 1
agile_v_skills ⭐ 38 stars
Agent Skills Library - The official collection of 20+ Agent Skills for the Agile V™ framework. Works with Claude Code, Cursor, VS Code, GitHub Copilot.
- 🎯 20+ specialized skills covering full product lifecycle
- 🔬 Formal traceability (REQ→ART→TC) prevents AI hallucinations
- 🛡️ Red Team verification catches what Build Agent misses
- 📋 Compliance-ready (ISO 9001, ISO 27001, GxP)
- 🚀 Multi-platform (Claude Code plugin, Cursor rules, VS Code skills)
agentic_agile_v ⭐ 1 star
Production-Ready Scaffold - Template for building verifiable AI-augmented engineering systems with structured briefs, evidence bundles, and CI gates.
- 🏗️ Project scaffold with automated traceability
- 📝 Structured briefs for requirements engineering
- 📊 Evidence bundles for verification
- 🔄 CI/CD integration with quality gates
- 🎯 TypeScript/Node.js template
| Feature | Typical AI Agents | Agile V Framework |
|---|---|---|
| Traceability | ❌ Code without requirements | ✅ Every artifact links to REQ-XXXX |
| Verification | ❌ Self-tests (confirmation bias) | ✅ Independent Red Team Verifier |
| Hardware | ❌ Assumes unlimited resources | ✅ Validates RAM/CPU/GPU constraints |
| Deployment | ❌ Autonomous production pushes | ✅ Human Gates with Evidence Summaries |
| Ambiguity | ❌ Silent assumptions | ✅ Halts and asks clarifying questions |
| Compliance | ❌ Manual audit prep (weeks) | ✅ Auto-generated ISO/GxP artifacts |
Claude Code:
/plugin install agile-v-skillsCursor:
cp -r .cursor/rules/ /path/to/your/project/.cursor/VS Code / GitHub Copilot:
cp -r agile-v-core/ ~/.copilot/skills/You: "Add user authentication"
Agent: ⚠️ HALT CONDITION: No requirement specification found
Invoking: requirement-architect
[Requirement Architect creates REQ-0001, REQ-0002...]
Build Agent: ✓ Implementation complete (ART-0001 → REQ-0001)
Red Team Verifier: Found 2 security issues
Build Agent: ✓ Issues fixed
Red Team Verifier: ✓ All tests pass
Agent: Ready for Human Gate approval
- 📖 Official Website — Framework overview and philosophy
- 📘 Agent Skills Guide — Complete skills documentation
- 📙 Examples — Before/after scenarios
- 📗 Skill Routing Guide — When to use which skill
- 📕 Compliance Matrices — ISO 9001, ISO 27001, GxP mappings
Ideal for:
- 🏭 Enterprise teams needing verifiable AI-augmented engineering
- 🏥 Regulated industries (medical devices, aerospace, pharma, finance)
- 🔬 R&D teams building mission-critical systems
- 🎓 Academic researchers studying AI safety and verifiability
- 🛡️ Security-conscious organizations requiring audit trails
Industry Coverage:
- Medical Devices (ISO 13485, 21 CFR Part 11)
- Aerospace (AS9100D)
- Pharmaceuticals (GxP, GAMP 5)
- Financial Services (SOC 2, GDPR)
- General Manufacturing (ISO 9001)
We welcome contributions! The Agile V™ Framework is open source (CC-BY-SA-4.0).
Ways to contribute:
- ⭐ Star our repos to show support
- 🐛 Report issues or suggest improvements
- 💡 Share use cases from your industry
- 🔧 Contribute new skills following our guidelines
- 📚 Improve documentation (especially examples)
- 🌍 Translate to other languages
- 🌐 Website: agile-v.org
- 📧 Email: info@agile-v.org
- 🐦 Twitter: @AgileVFramework
- 💼 LinkedIn: Agile V Framework
- 📖 GitHub: github.com/Agile-V
To make AI-augmented engineering verifiable, traceable, and trustworthy for mission-critical systems.
We believe AI agents should be powerful tools, not uncontrolled wildcards. The Agile V™ Framework provides the rigor and structure needed to build real products with AI assistance while maintaining the quality, traceability, and compliance requirements of enterprise and regulated environments.
All public Agile V™ projects are licensed under CC-BY-SA-4.0 (Creative Commons Attribution-ShareAlike 4.0 International).
This means:
- ✅ Free to use for commercial and personal projects
- ✅ Modify and adapt for your needs
- ✅ Share with attribution
⚠️ Derivatives must use the same license (ShareAlike)
Built with ❤️ by the Agile V™ Team