Anti-Infinite-Loop Guard
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The skill is a simple loop-prevention guard with no credentials, network access, persistence, or destructive behavior, though it can make an agent stop early and its safety claims are not fully substantiated by the included code.
This appears safe to install as a lightweight loop-prevention aid. Treat it as a simple heuristic guard rather than a fully validated watchdog, and test its thresholds before using it on important or long-running tasks.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may stop a task after fixed limits or repeated/no-progress actions, even if more work might have succeeded.
The skill deliberately instructs the agent to alter stopping conditions. That is the advertised safety purpose, but it directly affects whether a task continues or stops.
Use when: (1) detecting repeated actions, (2) enforcing termination conditions
Use it for loop-prone tasks and review or adjust limits before relying on it for long or complex work.
Users may overestimate how reliable or complete the loop guard is.
The documentation presents precise safety/performance metrics, but the included source and trivial tests do not substantiate those claims.
| Detection latency | <10ms |
Treat the performance and reliability metrics as unverified and test the guard in your own workflows before depending on it.