developer-hiring-intake

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill instructs the agent to execute local scripts (e.g., scripts/validate-intake.py, scripts/analyze-requirements.py) on generated JSON data. While these are part of the skill's defined workflow for validation and compliance, the source code for these scripts was not provided, preventing a full security audit of the execution logic.
  • [PROMPT_INJECTION] (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8) as it ingests untrusted user input to generate files and drive further logic.
  • Ingestion points: The skill accepts unstructured "hiring needs" from the user to populate JSON fields.
  • Boundary markers: Absent. There are no explicit instructions or delimiters provided to prevent the LLM from following instructions embedded within the user's role descriptions.
  • Capability inventory: The skill performs file system writes (intake-{role-title}-{date}.json) and executes Python scripts via the command line.
  • Sanitization: While the instructions mention "validation checks" for legal compliance and schema correctness, the actual sanitization logic resides in the unprovided scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:49 PM