Share
## https://sploitus.com/exploit?id=BEB33147-7848-5A56-B185-8F5DA8EE1BFC
# โก Vexa โ Autonomous Self-Healing Security Platform
> Continuously analyses repositories for vulnerable dependencies, validates exploits in isolated sandboxes, generates patches, and creates Pull Requests โ autonomously.
## Architecture
```
12 Agents across 3 swarms:
Blue Swarm (8 agents) Red Swarm (3 agents) Coordinator (1)
โโโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโ
Repo Analysis Technology Recon Decision Engine
Dependency Graph Endpoint Discovery
SBOM Exploit Validation
Threat Intelligence
Runtime Correlation
Patch Planning
Patch Generation
Test Validation
PR & Learning
```
## Quick Start
### 1. Configure environment
```bash
cp .env.example .env
# Fill in: OPENAI_API_KEY, GITHUB_TOKEN, GITHUB_CLIENT_ID, GITHUB_CLIENT_SECRET
```
### 2. Start all services
```bash
docker-compose up -d
```
Services:
- API: http://localhost:8000
- Dashboard: http://localhost:3000
- Docs: http://localhost:8000/docs
### 3. Scan a repository
**Dashboard:** Open http://localhost:3000, paste a GitHub URL, click Scan.
**CLI:**
```bash
python cli/vexa.py scan https://github.com/owner/repo
python cli/vexa.py report
python cli/vexa.py list
```
**API:**
```bash
curl -X POST http://localhost:8000/scans/ \
-H "Content-Type: application/json" \
-d '{"repo_url": "https://github.com/owner/repo"}'
```
**GitHub Actions:** Add `.github/workflows/vexa-scan.yml` to your repo and set:
- `VEXA_API_URL` secret
- `VEXA_API_KEY` secret
## Red Swarm (Authorized Testing Only)
Red Swarm runs exclusively inside an isolated Docker network with no external access.
Enable only with explicit authorization:
```bash
python cli/vexa.py scan https://github.com/owner/repo --red-swarm
```
## Outputs
| Output | Location |
|--------|----------|
| Security Report | `GET /reports/{scan_id}/security-report` |
| SBOM (CycloneDX) | `GET /reports/{scan_id}/sbom?format=cyclonedx` |
| SBOM (SPDX) | `GET /reports/{scan_id}/sbom?format=spdx` |
| GitHub PR | Automatically created on scan completion |
| Audit Log | Included in security report |
## Stack
| Layer | Technology |
|-------|-----------|
| Agent Framework | LangGraph + LangChain |
| Backend | FastAPI |
| Frontend | Next.js |
| Database | PostgreSQL |
| Cache + Events | Redis Streams |
| Scanner | Trivy |
| SBOM | Syft |
| Threat Intel | OSV + NVD + GitHub Advisory |
| Containers | Docker |
| Auth | GitHub OAuth |
| CI/CD | GitHub Actions |