— Legal & Contracts

NDA triage

Grades every clause in an incoming NDA against your saved house positions and returns a single verdict — sign, mark up (with the drafting done) or escalate.

  • Legal & Contracts
  • Deterministic scoring
  • Traceable reasoning
  • Runs anywhere
Preview methodology

30–45 minutes → ~3 minutes

For one complete, review-ready pass

No coding required

Input
Incoming non-disclosure agreements
Output
Overall sign, amend or escalate recommendation
Runs in
Skillsize · ChatGPT · Claude · Copilot
Export
SKILL.md · MCP
Time saved
~30–45 minutes per run

— USE CASES

What people use NDA triage for

Routine NDA review

Provides a clear recommendation on whether an incoming NDA aligns with the firm’s standard positions and can be signed as drafted.

Targeted amendments for negotiation

Identifies departures that can be addressed through limited changes and supplies proposed amendment wording grounded in house positions.

Terms requiring legal scrutiny

Flags NDAs that warrant escalation and explains which clauses drive that recommendation, giving legal advisers a focused basis for review.

Consistent review across teams

Applies the same house positions across incoming NDAs, reducing variation between reviewers and making the rationale for each recommendation explicit.

What it works from

  • Incoming non-disclosure agreements
  • The firm’s standard NDA positions
  • Precedent NDAs reflecting accepted house terms

What you get back

  • Overall sign, amend or escalate recommendation
  • Clause-level assessment against house positions
  • Draft amendment wording for manageable departures
  • Reasons for escalation and clauses requiring closer scrutiny

— HOW IT BEHAVES

How NDA triage produces its result

The mechanics behind this specific template — what it reads, what it calculates, and where a human stays in the loop.

Consistent classification of clauses

Clauses are sorted into your categories using the same rules each time, which is what makes a large volume of contracts and agreements readable.

Every clause graded against one saved standard

Each clause is measured against a calibrated standard you own, so the bar for the contract portfolio stays identical between runs, assessors and months.

Composed as work product

Findings on the contract portfolio are written up as a document that reads like professional output, with each claim tied back to a clause.

Contracts and agreements broken into reviewable clauses

Long material is split into individual clauses, so each one is assessed on its own merits instead of buried in a document-level verdict.

Why this is expensive by hand

NDA reviews can absorb disproportionate time when reviewers interpret familiar clauses differently or revisit positions the firm has already settled. Routine NDA review is the typical trigger — provides a clear recommendation on whether an incoming NDA aligns with the firm’s standard positions and can be signed as drafted. Done properly it is defensible; done at pace it becomes a judgement call nobody can retrace. And "properly" usually means 30–45 minutes of manual work.

How this Skill produces it

Here the same job runs as a Skill. Your material goes in; overall sign, amend or escalate recommendation comes out, alongside clause-level assessment against house positions. The judgement is built in — how items are broken up, what standard they are held to, and where the run stops for a human review. The practical effect: 30–45 minutes of manual work becomes a ~3 minutes run, held to an identical standard on the tenth engagement as on the first.

Who it's for

  • In-house legal and contract managers
  • Commercial and procurement leads reviewing terms
  • Operations teams handling contract volume
  • Consultants advising on contractual risk

Run it in Skillsize — or export it anywhere

NDA triage exports as a structured SKILL.md file and is MCP-ready, so the same method runs in ChatGPT, Claude, Copilot or your own AI products. Adapt it to your methodology, and the intelligence stays yours — not locked to one vendor.

ChatGPTClaudeCopilotAI Products (MCP)