charles-proxy-extract

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill processes untrusted network log data and outputs it to the agent, creating a surface for indirect instructions. \n
  • Ingestion points: extract_responses.py (line 42) reads data from user-provided files using json.load(). \n
  • Boundary markers: Absent. The script prints extracted text without delimiters or instructions to ignore embedded content. \n
  • Capability inventory: Limited to local file operations; no internal network or shell execution capabilities. \n
  • Sanitization: Content is extracted and displayed raw without sanitization or escaping.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 01:13 PM