cpp
Pass
Audited by Gen Agent Trust Hub on May 7, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The SKILL.md file includes a 'Memory Protocol' that directs the agent to execute a shell command (
cat .claude/context/memory/learnings.md) to restore context between sessions. This is a functional requirement for the intended development workflow but involves direct shell interaction. - [PROMPT_INJECTION]: The skill is designed to process and review user-provided source code while having access to sensitive tools, creating an indirect prompt injection surface where instructions hidden in code comments could attempt to manipulate the agent.
- Ingestion points: Processes files matching common C++ and build system globs (e.g., .cpp, .h, CMakeLists.txt) as defined in SKILL.md and schemas/input.schema.json.
- Boundary markers: No specific delimiters or instructions to ignore embedded natural language directives were identified in the guidelines.
- Capability inventory: The skill is granted access to high-privilege tools including
Read,Write,Edit,Bash,Grep, andGlob. - Sanitization: There is no evidence of input sanitization to filter or escape natural language instructions embedded within the processed code files.
Audit Metadata