skills/suzheng/k2f/k2f-contract/Gen Agent Trust Hub

k2f-contract

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of a companion tool using the command npx skills add suzheng/k2f. This fetches additional functionality from the vendor's repository to handle contract packaging.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data such as deal notes, emails, and user-provided terms to generate contract blueprints.
  • Ingestion points: User-supplied deal notes, emails, and commercial terms defined in SKILL.md.
  • Boundary markers: The instructions mandate labeling missing information as NEEDS USER CONFIRMATION and using placeholders for unsupplied data.
  • Capability inventory: This skill focuses on blueprint generation; the actual packaging and execution logic reside in the referenced k2f skill.
  • Sanitization: Implements a 'Counsel gate' that emphasizes output is not legal advice and an 'Honesty gate' to prevent the agent from fabricating legal facts or registrations.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 12:21 PM