OpenA2A builds open-source tools for securing AI agents in production. AI agents are the fastest-growing category of non-human identities, and most organizations have no governance, no visibility, and no security controls around them. We're building the infrastructure to fix that -- identity management, runtime protection, security scanning, compliance benchmarks, and credential management for AI developer tools. Everything is Apache-2.0, self-hostable, and designed to work independently or together.
| Project | Description | Install |
|---|---|---|
| AIM | Identity & access management for AI agents | pip install aim-sdk |
| HackMyAgent | Security scanner -- 147 checks, attack mode, auto-fix | npx hackmyagent secure |
| OASB | Open Agent Security Benchmark -- 182 attack scenarios | npm install @opena2a/oasb |
| ARP | Agent Runtime Protection -- process, network, filesystem monitoring | npm install @opena2a/arp |
| Secretless AI | Credential management for AI coding tools -- Claude Code, Cursor, Windsurf | npx secretless-ai init |
| DVAA | Deliberately vulnerable AI agents for security training | docker pull opena2a/dvaa |
┌──────────────────────────────────────────────────────────┐
│ Your AI Agent │
│ │
│ Secretless AI → Credential management for dev tools │
│ AIM → Identity, governance, access control │
│ ARP → Runtime process/network/file monitoring │
│ HackMyAgent → Scan, harden, attack-test │
│ OASB → Compliance benchmark (46 controls) │
│ DVAA → Train your team on AI agent security │
└──────────────────────────────────────────────────────────┘
We contribute security fixes back to the open-source projects we depend on and audit.
OpenClaw -- 8 security PRs (2 merged, 6 open):
- Credential redaction in gateway config responses (#9858)
- Skill/plugin code safety scanner (#9806)
- Path traversal prevention in A2UI file serving (#10525)
- Security headers for gateway HTTP responses (#10526)
- Timing-safe comparison for hook token auth (#10527)
- Supply chain hardening with --ignore-scripts (#10528)
- File permission enforcement for credential files (#10529)
- Skill scanner false positive reduction (#10530)
Nanobot -- 1 security PR (open):
- Path traversal, XSS, and shell escape fixes (#472)
All projects are Apache-2.0.