Homepage

agent-threat-rules

Open detection standard -- like Sigma, but for AI agents. 683 rules for prompt injection, tool poisoning, context exfiltration, and MCP attacks. Shipped in Cisco AI Defense. 97.2% recall on NVIDIA garak.

< Back for full overview of agent-threat-rules vulnerabilities

Package versions

50 VERSIONS IN TOTAL
versionpublisheddirect vulnerabilities
2.0.188 May, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.1721 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.1621 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.1521 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.1421 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.1321 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.1221 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.1121 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.1021 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.919 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.819 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.719 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.619 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.518 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.318 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.216 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.115 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
2.0.015 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
1.2.010 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L
1.1.18 Apr, 2026
  • 0
    C
  • 0
    H
  • 0
    M
  • 0
    L

Automatically find and fix vulnerabilities affecting your projects. Snyk scans for vulnerabilities and provides fixes for free.

Get started free