add-doc-updater

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly requires a DOC_SOURCE_URL and the implementation/template (see Phase 0 variable DOC_SOURCE_URL and the "Download & Extract" section in references/doc-updater-template.md) to download and ingest upstream public archives of markdown which are post-processed for "AI consumption" and then used by agents in the Phase 5 integration test, so untrusted third‑party content can directly influence agent behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The generated updater script downloads the provided DOC_SOURCE_URL at runtime (e.g., https://gitlab.com/gitlab-org/cli/-/archive/main/cli-main.tar.gz?path=docs) and injects the fetched markdown into the skill/agent context for AI consumption, so external content can directly control prompts.