Embedded C Code Review
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for static analysis of C and C++ code. Its instructions correctly focus on identifying technical issues such as race conditions, volatile correctness, and naming convention compliance.
- [DATA_INGESTION]: The skill ingests untrusted source code provided by the user for review. This represents a surface for Indirect Prompt Injection (Category 8). However, the skill provides specific structured templates for output and focuses on analysis rather than execution, which significantly mitigates risk. Boundary markers are partially addressed by the 'Operating principles' that guide the agent to work within a specific review scope.
- [COMMAND_EXECUTION]: The skill instructions include logic to detect a project root by searching for a '.git' directory and writing results to an 'analysis/' folder. These are standard operations for a local developer tool and are not used maliciously.
- [NO_REMOTE_CONTENT]: The skill utilizes local reference files ('adaptive-ap.md' and 'html-report-template.html') for its functionality. No external scripts or packages are downloaded or executed at runtime.
- [METADATA_INTEGRITY]: The skill's metadata and descriptions accurately reflect its functionality as a senior embedded engineer's code review tool.
Audit Metadata