sales-convin

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to WebFetch other skills' SKILL.md from the public raw GitHub URL (https://raw.githubusercontent.com/sales-skills/sales/main/skills/{skill-name}/SKILL.md) and to read those files as part of its routing/decision workflow, which exposes it to untrusted public repo content that could alter actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs the agent at runtime to WebFetch and read SKILL.md from https://raw.githubusercontent.com/sales-skills/sales/main/skills/{skill-name}/SKILL.md, a required remote fetch whose contents are used to control routing and response prompts.