gamma
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a bash script
scripts/gamma.shto executecurlandjqfor API communication and JSON manipulation.\n- [DATA_EXFILTRATION]: The skill transmits user-provided text content and a sensitive API key to thepublic-api.gamma.appdomain. This is consistent with the skill's primary purpose and targets a well-known service endpoint.\n- [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by acting as a conduit for untrusted data to a downstream AI service.\n - Ingestion points: User-supplied content or contents of local files passed as arguments to the
generatecommand inscripts/gamma.sh.\n - Boundary markers: No delimiters or instructions are used to distinguish user data from agent instructions in the API payload.\n
- Capability inventory: The skill utilizes
curlfor network requests and is documented to handle file-based input.\n - Sanitization: While the skill uses
jqto ensure valid JSON structure, it lacks semantic validation or sanitization of the natural language content within the payload.
Audit Metadata