Commute Matrix

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection when processing external data from camp provider files. If these files contain malicious instructions, the agent might follow them when generating summaries or recommendations.\n
  • Ingestion points: Provider markdown files located in <research_dir>/providers/*.md, as processed in SKILL.md (Step 2) and scripts/commute_calculator.py (scan_providers function).\n
  • Boundary markers: The skill instructions do not define delimiters or provide 'ignore embedded instructions' directives to isolate external content.\n
  • Capability inventory: The agent is tasked with generating recommendations and updating local markdown files based on the processed data, as seen in SKILL.md (Steps 4 and 5) and scripts/commute_calculator.py (update_provider_files function).\n
  • Sanitization: While the Python script uses regex to extract specific fields, it does not sanitize the resulting strings to prevent them from being interpreted as instructions by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 12:01 AM
Security Audit — agent-trust-hub — Commute Matrix