ida-mcp-headless-reverse-engineering

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime workflow ingests outsider-authored free text when the operating user points it at an arbitrary binary file (e.g., open_idb(path=...) / open_dsc(path=...)), because the LLM context can include extracted strings/decompiled pseudocode/disassembly originating from that outsider-provided file’s contents.