lovstudio-fill-web-form

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted content from external URLs and processes it using a high-privilege sub-agent without proper sanitization.\n
  • Ingestion points: Data enters the system via WebFetch in SKILL.md (Step 1) from user-provided URLs.\n
  • Boundary markers: No explicit delimiters or instructions are used to isolate the external content from the sub-agent's primary instructions.\n
  • Capability inventory: The skill has access to WebFetch, Read, Write, Glob, Grep, and a sub-agent with exploration capabilities.\n
  • Sanitization: The skill does not filter or sanitize the extracted form fields before passing them to the sub-agent in Step 2, creating a vector for cross-domain data access or manipulation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 02:54 PM
Security Audit — agent-trust-hub — lovstudio-fill-web-form