agreement-generator

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed for text and document generation based on user-provided parameters. It limits its operation to a safe subset of tools (Read, Write, Glob, Grep) and does not request shell access or network capabilities.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data during its information-gathering wizard to populate agreement templates.
  • Ingestion points: User-provided names, addresses, and business terms collected during the 'information-gathering wizard' (SKILL.md).
  • Boundary markers: Absent; the instructions do not specify delimiters or warnings for the agent to ignore embedded instructions within user inputs.
  • Capability inventory: The skill uses 'Write', 'Read', 'Glob', and 'Grep' tools to manage the generated agreement files.
  • Sanitization: Absent; no explicit validation or escaping of user input is described.
  • Note: While the surface exists, the lack of dangerous capabilities (like shell execution) and the descriptive nature of the task render this risk minimal.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 09:29 PM