article-depth-summarizer

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection (LOW): The skill is susceptible to indirect prompt injection as it ingests untrusted data from the web and utilizes file-writing capabilities. * Ingestion points: The skill uses read_url_content to fetch content from user-provided URLs. * Boundary markers: Absent; there are no instructions to the agent to treat the fetched content purely as data or to ignore embedded instructions. * Capability inventory: The skill uses the write_to_file tool to save reports to the local file system. * Sanitization: Absent; there is no explicit instruction to sanitize fetched content before processing or to prevent path traversal in the generated filename.
  • Metadata Poisoning (SAFE): The metadata accurately describes the skill's purpose and does not contain deceptive instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:39 PM