infra-usage-analyzer
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external structured data (JSON/CSV) which could potentially contain malicious instructions intended to influence the agent's behavior.
- Ingestion points: The
run.pyscript reads data from a file path provided via the--inputcommand-line argument usingPath.read_text(). - Boundary markers: There are no specific delimiters or instructions provided to the agent to distinguish between its own logic and data-embedded instructions.
- Capability inventory: The script has the capability to read files (
Path.read_text) and write files (Path.write_text) to the filesystem, and it can create directories usingmkdir(). - Sanitization: The input is parsed as JSON using
json.loads(), which provides structural validation but does not filter or sanitize the actual content of the data for natural language instructions.
Audit Metadata