deliverable-upgrade

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

• Third-party content exposure detected (high risk: 0.90). The skill fetches and compares a remote VERSION from raw.githubusercontent.com and may git-clone/git-fetch from https://github.com/canhta/deliverable.git as part of its required upgrade flow, so it ingests untrusted public third-party content that directly determines whether and how it updates the agent's code.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly runs a runtime git clone of https://github.com/canhta/deliverable.git to replace the local install, which fetches remote code that will become executable skill code (remote code execution risk).