roblox-docs

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

• Third-party content exposure detected (high risk: 0.85). The required workflow clones and reads a public GitHub repository’s markdown/YAML files (outsider-authored free text) at runtime, and then the subagent returns that file content/snippets into the agent’s LLM context for summarization.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's script (scripts/robloxdocs.py) performs runtime git clone/pull from https://github.com/Roblox/creator-docs and injects the fetched docs into the agent's search/read flow to produce answers, so the external repo directly controls agent outputs.