gemini-cli-docs

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md explicitly requires scraping and ingesting public docs from https://geminicli.com/llms.txt (via scripts/core/scrape_docs.py and manage_index.py) and then uses that fetched content to drive searches, section extraction, index rebuilds and drift/cleanup actions, so external web content can materially influence agent behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly runs scraping scripts at runtime that fetch and ingest external documentation from https://geminicli.com/llms.txt, which can be injected into the agent's context and thus directly influence prompts/responses (runtime fetch + content-in-context + required for scraping workflows).