agentforce-architecture-analyze

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text is ingested into the LLM context via runtime-fetched Salesforce metadata bodies (e.g., Flow Metadata complexvalue, Apex Body/SymbolTable, and GenAi PromptTemplate text) that are retrieved from the target org over Tooling/Data APIs and then rendered into architecture.md/metadata_tree.json for the agent to read.