reality-validation

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, presenting an indirect prompt injection surface.
  • Ingestion points: Production data samples, historical transaction data, and user observations listed in SKILL.md.
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are provided for the agent when processing this data.
  • Capability inventory: The skill utilizes subprocess execution through npm scripts and shell commands as described in SKILL.md and scripts/example.sh.
  • Sanitization: The instructions explicitly recommend the use of sanitized production data in SKILL.md.
  • [DATA_EXFILTRATION]: The core functionality of the skill requires accessing and analyzing potentially sensitive data sources like patient schedules or financial records. While intended for validation, this ingestion path creates a surface for data exposure if not handled within a secure environment.
  • [COMMAND_EXECUTION]: The skill contains a shell script (scripts/example.sh) and provides numerous examples of npm command execution for running validation tasks. These are standard actions for the defined purpose and do not show signs of malicious intent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 02:45 PM