synthesis-llm-setup

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

• Third-party content exposure detected (high risk: 0.90). This skill's workflow explicitly instructs syncing a GitHub repository to Claude and uploading/using all-knowledge.md as project knowledge (see "Claude Projects setup" and "ChatGPT GPT setup" / "Knowledge files"), so the agent will ingest third-party repository content that can contain untrusted, user-generated instructions that could alter its behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). Claude's GitHub sync will cause Claude to index and use all-knowledge.md from an external GitHub repository at runtime (e.g., https://github.com/your-org/ai-knowledge), meaning remote repo content can directly control the agent's prompts/instructions.