job-description-generator
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a highly structured text-generation task with no evidence of malicious patterns such as remote code execution, credential exfiltration, or persistence mechanisms.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from user-provided files (via
--comp-band,--essential-functions, andupdatearguments). While this is an attack surface for indirect prompt injection, the skill mitigates this risk by enforcing a strict 'source-faithful language' rule (Step 3) and requiring grounding in specific user inputs, which limits the model's susceptibility to instructions embedded in the data. - [COMMAND_EXECUTION]: The skill does not contain any shell command execution or subprocess calls. All operations are related to internal text processing and logic checks based on provided metadata.
- [DATA_EXPOSURE_AND_EXFILTRATION]: No network operations or sensitive file access patterns (e.g., SSH keys, AWS credentials) were identified. The file paths mentioned are restricted to job description drafts and configuration files within the task's expected context.
Audit Metadata