delegate

Fail

Audited by Snyk on Nov 18, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E004: Prompt injection detected in skill instructions.

  • Potential prompt injection detected (medium risk: 0.30). The skill is a delegation and orchestration guide with no sign of intent to attack the user; the one borderline item is that it tells the driver to always pass --permission-mode auto to spawned agents, which may loosen permission checks on the user's resources beyond what each task needs, so it is flagged as medium under the recall-first rule.

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

  • Third-party content exposure detected (low risk: 0.10). The skill's delegation workflow can send a delegated agent to browser capture and broad searches (SKILL.md lines 26 and 111), which read arbitrary web pages and search results the agent selects, so outsider text is reachable only after the agent actively searches or fetches it, and the skill does not itself require that.

Issues (2)

E004
CRITICAL

Prompt injection detected in skill instructions.

W011
MEDIUM

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

Audit Metadata
Risk Level
CRITICAL
Analyzed
Nov 18, 2026, 05:34 PM
Issues
2