glim-mcp
Warn
Audited by Snyk on Aug 20, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The skill’s runtime path is explicitly an on-demand research workflow where the outsider-authored text comes from platform feeds/pages the user requests to search and then the agent may perform “full-page extraction” on those results (web search + extraction; X/Twitter, Reddit, GitHub, Amazon, YouTube transcripts), and those fetched contents are treated as untrusted data.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata