c-systems
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted C source code and headers, which represents an attack surface for indirect prompt injection if the source files contain instructions disguised as comments or string literals.
- Ingestion points: User-provided C source and header files as defined in the paths frontmatter of SKILL.md.
- Boundary markers: The skill does not define explicit delimiters or warnings for the code being analyzed to prevent the agent from interpreting embedded content as instructions.
- Capability inventory: The agent uses compilers (gcc, clang), build systems (make, phpize), and analysis tools (valgrind, sanitizers), and is expected to perform code refactoring.
- Sanitization: The instructions focus on C-level safety but do not include specific prompt-level sanitization for the input code.
Audit Metadata