planning-with-files

Warn

Audited by Snyk on Mar 10, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.90). This skill's workflow explicitly requires performing web/browser/search operations and recording "Visual/Browser Findings" (see SKILL.md's "The 2-Action Rule", the "Visual/Browser Findings" section in templates/findings.md, and examples showing "WebSearch") so it ingests open/public web content into planning files that the agent rereads and uses to make decisions, enabling indirect prompt-injection from untrusted third-party sources.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 10, 2026, 04:39 PM