init-greptile-review

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

• Third-party content exposure detected (high risk: 0.80). The skill's docs and workflow explicitly permit and instruct Greptile to index and ingest external repositories and context files (e.g., the "patternRepositories" org/repo setting, .greptile/files.json entries, and "full codebase indexing" in references/setup.md), meaning the agent will read user-generated content from other public GitHub/GitLab repos which can be used as review context and therefore could indirectly inject instructions that influence review decisions.