snap-slice

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md Workflow step 1 explicitly requires reading the PRD issue body and all comments, extracting refs/URLs and recursing into referenced issues/PRs/discussions/docs (i.e., user-generated or third-party web content) which the agent must interpret to decide slices, blockers, and next actions, exposing it to untrusted indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly reads referenced GitHub issues/PRs/docs at runtime (e.g., https://github.com/OWNER/REPO/issues/10 and https://github.com///commit/) and uses that fetched content to drive prompt generation and slicing logic, so these external URLs can directly control the agent's instructions.