writing-skills

Warn

Audited by Snyk on Feb 27, 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.md (see "Package dependencies" and "Runtime environment" in anthropic-best-practices.md) explicitly allows the agent to install packages from npm/PyPI and pull from GitHub repositories, meaning the agent may fetch and execute untrusted, user‑generated content from public third‑party sources which can materially influence tool use or next actions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 27, 2026, 08:48 AM