skill-creator

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to "Fetch the URL content (curl or web-fetch skill)" in the "Installing a Skill from URL" workflow and to extract and act on the fetched SKILL.md (including executing any defined installation steps or scripts), meaning arbitrary public URLs and their user-provided instructions would be read and could change the agent's subsequent actions.