geo-optimizer

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves data from external URLs via web_fetch and processes user-pasted content without implementing boundary markers or instructions to disregard potential commands within that data.
  • Ingestion points: Step 1 utilizes the web_fetch tool to retrieve content from arbitrary user-provided URLs and also accepts raw text pasted by the user.
  • Boundary markers: None. The instructions do not specify the use of delimiters or provide warnings to the agent to ignore instructions embedded within the fetched or pasted material.
  • Capability inventory: The skill has the ability to perform network reads via web_fetch and write files (HTML and DOCX) to the /mnt/user-data/outputs/ path.
  • Sanitization: There are no mentioned processes for sanitizing, escaping, or validating the input data before it is processed by the agent for rewriting.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 08:04 AM
Security Audit — agent-trust-hub — geo-optimizer