privacy-policy-diff-summarizer

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted text from external files which are then summarized by the agent. This represents an indirect prompt injection surface where instructions hidden within a privacy policy could influence the agent's output.\n
  • Ingestion points: The privacy_policy_diff_summarizer.py script reads data from file paths provided via the --old and --new arguments.\n
  • Boundary markers: Absent. The script extracts snippets from the text and includes them in a JSON report without specific delimiters or warnings to the agent to ignore embedded instructions.\n
  • Capability inventory: The script performs local file reads and outputs results to stdout; it does not have network or elevated system capabilities.\n
  • Sanitization: The script normalizes whitespace but does not filter or sanitize the text content for potential malicious prompt sequences.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 04:09 PM
Security Audit — agent-trust-hub — privacy-policy-diff-summarizer