tinyfish-social-listening

Warn

Audited by Snyk on Jul 6, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.95). The skill’s runtime workflow uses TinyFish Search to ingest outsider-authored web content snippets (e.g., Reddit posts, news articles, forum threads, tweets) into the agent’s context for date-gating and then fetching/summarizing them, via the tinyfish search query ... → snippet parsing → tinyfish fetch content get ... path.

Issues (1)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 6, 2026, 07:56 AM
Issues
1
Security Audit — snyk — tinyfish-social-listening