fastly-cli

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

• Third-party content exposure detected (high risk: 0.80). The skill's compute initialization workflow (references/compute.md) explicitly accepts Git repository and archive URLs (e.g., "fastly compute init -f https://github.com/…"), which causes the agent/CLI to fetch and ingest arbitrary public repo contents (including fastly.toml, build scripts, and templates) that can influence subsequent build/deploy actions — i.e., untrusted third-party content is read and can materially change behavior.