ip-clause-review

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection Surface: The skill is designed to analyze untrusted text from legal agreements. As with any system processing external data, there is a potential for embedded content to attempt to influence the AI's output.
  • Ingestion points: Legal agreement content provided via file paths, Drive links, or pasted text as described in SKILL.md.
  • Boundary markers: The skill does not currently specify explicit delimiters or instructions to ignore embedded commands within the agreement text.
  • Capability inventory: The skill has the ability to read practice profiles, write reports to local matter folders within the plugin directory, and interact with configured legal research tools (SKILL.md).
  • Sanitization: The instructions do not define specific sanitization or filtering protocols for the input text prior to analysis.
  • Local Workspace Access: The skill reads and writes to specific paths within the ~/.claude/ directory to manage practice profiles and matter-specific context. This allows for personalized analysis according to the user's defined legal standards and project workspaces.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 09:46 PM
Security Audit — agent-trust-hub — ip-clause-review