github

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

• Third-party content exposure detected (high risk: 0.80). This skill calls the public GitHub API (fetching from https://api.github.com in src/api.ts / dist/api.js and is described in SKILL.md/README) to retrieve user-generated repository, commit, issue and CI data which the agent is expected to read and use as part of its workflow (list_repos, get_recent_activity, check_ci_status, search_repos), so untrusted third‑party content can influence subsequent actions.