win64-native-builds
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to compile C/C++ source code into native Win64 executables using Clang or GCC and to run the resulting binaries as part of a test suite.
- [DYNAMIC_EXECUTION]: In the 'Hermes terminal discipline' section, the skill explicitly recommends a pattern of writing complex logic to a shell script (.sh) using file writing tools and then executing that script via a terminal command.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted local data (source code and test files) and uses it to drive terminal operations, creating a surface for indirect prompt injection if the source files contain malicious instructions.
- Ingestion points: Local source files (e.g., main.c, *_test.c) and version files.
- Boundary markers: No specific boundary markers or 'ignore embedded instructions' warnings are provided for the data being read.
- Capability inventory: The skill utilizes terminal execution, file system writes (write_file), and binary execution.
- Sanitization: No explicit sanitization or validation of the source code content is described before it is used in build commands or scripts.
Audit Metadata