Nda Triage

What it does

This skill enables AI agents to analyze incoming requests against a repository of Non-Disclosure Agreements (NDAs) to determine if specific information is protected. It automatically identifies potential conflicts between user inputs and existing confidentiality clauses before processing the data further.

When to use it

• Pre-processing security checks: Before an agent begins drafting legal documents or analyzing sensitive client data, run this skill to ensure no NDA violations occur.
• Automated compliance filtering: Integrate into a workflow where agents must route flagged content to human legal review instead of executing tasks automatically.
• Cross-company collaboration: Use when multiple AI agents from different organizations need to exchange information while adhering to strict confidentiality boundaries.
• Audit trail generation: Leverage the skill to log which specific NDA clauses were triggered by user inputs for future compliance reporting.

Key capabilities

• Scans user-provided text against a defined set of NDA documents.
• Identifies specific clauses that may be violated by the input content.
• Returns structured data indicating whether the information is safe to process or requires human intervention.
• Supports dynamic updates to the NDA repository as new agreements are signed.

Example prompts

• "Analyze this client's project brief against our current NDAs and flag any sensitive details that might be restricted."
• "Check if the attached research data violates Clause 4 of the signed agreement before proceeding with analysis."
• "Run an NDA triage on this email draft to ensure no confidential information is accidentally disclosed to external parties."

Tips & gotchas

Ensure your NDA repository is up-to-date, as the skill relies entirely on the documents currently loaded in the agent's context. The tool identifies potential conflicts but does not provide definitive legal advice; always have a human review flagged items before finalizing actions.