report

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from the input/ directory to generate reports, creating an indirect prompt injection surface.
  • Ingestion points: Untrusted data is read from files in input/<dataset>/ using the Read and Glob tools as specified in the instructions.
  • Boundary markers: The report assembly instructions (Step 4) do not define delimiters or instructions for the agent to ignore embedded commands within the processed data.
  • Capability inventory: The skill has access to Bash, Write, and Edit tools, which provide a high-impact capability set if the agent inadvertently follows instructions found within data files.
  • Sanitization: There is no evidence of data validation, escaping, or filtering before interpolation into the Markdown report templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 08:51 AM
Security Audit — agent-trust-hub — report