mcu-embedded-review
Fail
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: CRITICALINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external firmware source code during the review process, creating a surface for indirect prompt injection where malicious instructions could be embedded in code comments or documentation to influence the agent. 1. Ingestion points: Firmware source code files accessed via Read and Grep tools. 2. Boundary markers: None; the skill does not instruct the agent to ignore instructions embedded within the data. 3. Capability inventory: The configuration allows Bash(cargo:build), providing a method to execute code during compilation. 4. Sanitization: No specific sanitization or filtering is applied to the ingested source code content.
- [DATA_EXFILTRATION]: The documentation contains a hardcoded absolute file path (/Users/cam/Developer/muni/CLAUDE.md) which reveals a local system username ('cam') and the author's internal directory structure.
Recommendations
- CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
Audit Metadata