subagent-teams

Warn

Audited by Snyk on Feb 22, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.80). The skill explicitly grants Explore/Plan subagents the WebFetch and WebSearch tools (see Task Tool Parameters and reference/task-tool-guide.md) and includes prompt templates like "Framework Evaluation" that instruct agents to research external frameworks and return GO/NO-GO recommendations, meaning agents will fetch and interpret untrusted public web content that can materially influence decisions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 22, 2026, 09:04 PM