AI security scanner for OpenClaw - powered by AgentTinman. Discovers prompt injection, tool exfil, context bleed, and other security issues in your AI assistant sessions, then proposes mitigations mapped to OpenClaw's security controls.
Stars
3
Forks
1
Watchers
3
Open Issues
0
Overall repository health assessment
No package.json found
This might not be a Node.js project
17
commits
Release v0.6.2 with safer gateway defaults and event redaction
9d5f336View on GitHubUpdate SKILL.md to v0.6.0 with check, mode, allow commands
51e3799View on GitHubRestore all detection patterns: 168 patterns across 16 categories
37340f0View on GitHubv0.6.0: Add security check system with safer/risky/yolo modes
cf74c24View on GitHubUpdate README for v0.5.1 with init command and watch stop
78b05e9View on GitHubAdd init command, fix tool args bug, implement watch stop
71c2119View on GitHubAdd detection patterns for evasion, platform attacks, memory poisoning
b22a97bView on GitHubv0.4.0: Enhanced security detection with 180+ attack probes
cb8bde5View on GitHub