fleet-safety

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8). It ingests a wide range of external data, including driver records, accident reports, and regulatory documentation, which could contain malicious instructions.\n
  • Ingestion points: The skill processes driver records, accident reports, violation logs, and telematics event data from the local file system (SKILL.md).\n
  • Boundary markers: There are no explicit delimiters or instructions to the agent to ignore embedded commands within the ingested data, which may allow malicious content in safety records to influence agent behavior.\n
  • Capability inventory: The skill utilizes file system read and write capabilities to generate reports in the docs/ directory and maintain telemetry logs.\n
  • Sanitization: No input validation or sanitization routines are implemented for the external data sources prior to analysis.\n- [SAFE]: The telemetry logging to hidden project directories (~/.claude/projects/) is a local persistence mechanism used for session tracking and self-improvement, which is consistent with the targeted platform's architecture and does not involve unauthorized network activity or data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 10:57 AM
Security Audit — agent-trust-hub — fleet-safety