matlab-import-export-vehicle-data
Pass
Audited by Gen Agent Trust Hub on Aug 2, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted vehicle network database and log files which serves as a potential surface for indirect prompt injection attacks.
- Ingestion points: The skill uses
fileread()inSKILL.md,mdfRead()inreferences/mdf-import.md,blfread()inreferences/blf-import.md, andcanSignalImport()inreferences/asc-import.mdto ingest external data. - Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following potential instructions embedded within the metadata or signal names of the processed files.
- Capability inventory: The skill utilizes the
evaluate_matlab_codetool as specified inmanifest.yaml, which executes MATLAB code generated or influenced by the processed data. - Sanitization: No sanitization or validation logic is present to filter signal names, message names, or comments from the external files before they are interpolated into the agent's context or code.
Audit Metadata