find-communities

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted content from external websites and processes it to create local files.
  • Ingestion points: Untrusted data enters the agent context through the WebSearch and WebFetch tools while surveying community platforms (e.g., LinkedIn, Discord, Slack).
  • Boundary markers: The instructions do not specify any delimiters or safety warnings to the agent regarding potential instructions hidden within the scraped web content.
  • Capability inventory: The skill utilizes the Write tool to generate and save markdown files (communities.md) to the local filesystem.
  • Sanitization: There is no mention of sanitizing or escaping the data retrieved from external sources before it is written to the output file.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 05:16 AM
Security Audit — agent-trust-hub — find-communities