moai-lang-cpp
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill explicitly allows the 'Bash' tool. While necessary for compiling C++ code and running tests, this tool provides the agent with the capability to execute arbitrary shell commands on the host environment.\n- [PROMPT_INJECTION] (LOW): Category 8: Indirect Prompt Injection surface. The skill is designed to ingest and process external C++ source code and build configuration files which may contain adversarial instructions. 1. Ingestion points: .cpp, .hpp, .h, CMakeLists.txt, vcpkg.json, conanfile.txt. 2. Boundary markers: Absent. 3. Capability inventory: Bash, Read, Grep, Glob, mcp__context7__resolve-library-id, mcp__context7__get-library-docs. 4. Sanitization: Absent.\n- [EXTERNAL_DOWNLOADS] (SAFE): The skill references industry-standard package managers and libraries including vcpkg (Microsoft), Conan, and Google Test. As Microsoft and Google are trusted organizations, the associated download and bootstrapping patterns (e.g., git clone, bootstrap-vcpkg.sh) are considered safe and downgraded to LOW/INFO per the trust rules.
Audit Metadata