5-5-3-2

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

• Third-party content exposure detected (high risk: 1.00). The skill's required "Auto-Update Check" (SKILL.md step 1) mandates running scripts/check_update.py, which fetches and parses a remote SKILL.md from a raw GitHub URL (https://raw.githubusercontent.com/...) and can cause the agent to update/reload skill instructions—exposing it to untrusted, user-controlled third‑party content that can influence behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The required runtime script scripts/check_update.py fetches and parses a remote SKILL.md from the constructed URL pattern "https://raw.githubusercontent.com/{owner}/{repo}/main/.agents/skills/{skill_name}/SKILL.md" and will cause the agent to run the supplied UPDATE_COMMAND (e.g., "npx skills update {owner}/{repo} --skill {skill_name}"), which can pull and execute remote code or new instructions, so this is a high-confidence runtime dependency that can control the agent.