ask-adr-logger
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The
create_adr.pyscript accepts an arbitrary directory path via the--dirargument without validation or restriction to a safe workspace. This creates an attack surface where a malicious prompt could influence the agent to write files to sensitive locations on the filesystem.\n - Ingestion points: Command-line arguments
--dirand--titleinscripts/create_adr.py.\n - Boundary markers: Absent. There are no instructions or delimiters to prevent the agent from accepting and using malicious paths.\n
- Capability inventory: File system write access via
os.makedirs()andopen(..., 'w')inscripts/create_adr.py.\n - Sanitization: While the
slugify()function sanitizes the filename portion, the directory path provided to the script is used directly without validation.
Audit Metadata