planning-with-files

Warn

Audited by Snyk on Jul 28, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (medium risk: 0.65). This skill’s workflow includes reading/writing notes.md (and task_plan.md) as “working memory,” and the provided examples explicitly show using WebSearch to gather sources and then writing those findings into notes.md, which—if those sources are fetched from the public web—would be outsider-authored free text that the LLM later reads from notes.md during “Read notes.md”.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 28, 2026, 08:50 AM
Issues
1