sprint

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md Step 2 (“Fetch Requirements”) instructs issue-worker subagents to “Read GitHub issue #67/#68” and extract goal/acceptance criteria/files likely to be touched, which is outsider-authored free text from public/authenticated GitHub issue bodies that is then embedded into the teammates’ prompts in Step 5.