> ## Documentation Index
> Fetch the complete documentation index at: https://vaquill.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# NDA Bulk Review
> Upload one or many NDAs and get a portfolio risk dashboard, per-document benchmark scoring, and concrete redline suggestions
**Best for:** pre-signing review across 5 to 50 NDAs in one upload, with traffic-light risk tiers and concrete redline language ready to paste.
Upload one or many NDAs and receive a portfolio risk dashboard, per-document benchmark scoring, and concrete redline suggestions. The point is volume: reviewing forty counterparty NDAs one at a time is a week; this is one launch and a coffee.
| | |
| -------------- | ---------------------------------- |
| Category | In-House |
| Jurisdiction | Any |
| Practice areas | Commercial, general, transactional |
| Runtime | 3-12 minutes |
## What you need
| Slot | Required | Count | What goes in it |
| ------------------ | -------- | ----- | ---------------------------------------------------- |
| **NDAs to review** | Yes | 1-500 | The NDAs under review. PDF or DOCX, text-searchable. |
| **Your playbook** | No | 0-1 | An optional company NDA playbook to compare against. |
## Inputs
| Field | Type | Required | What it changes |
| --------------------------------------------- | ------ | -------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Matter / engagement name | Text | No | Used in the matrix and DOCX artifact titles. |
| **Which side are you on?** | Select | **Yes** | Receiving Party · Disclosing Party · Mutual / Either Side. Determines what counts as risky. This inverts most of the scoring rubric, so get it right. |
| Your industry | Select | No | SaaS/Software, Fintech, Healthcare/HIPAA, Biotech, Hardware, AI/ML, Consumer/Retail, Government/Defense, Professional Services, Other. Calibrates residuals tolerance, trade-secret definitions, and which carve-outs (HIPAA, GLBA, BAA) count as required. |
| Purpose of disclosure | Select | No | M\&A diligence, vendor evaluation, customer pilot, partnership, hiring, investor, other. Purpose creep is a top NDA red flag; setting this lets the review flag mismatches between the stated purpose and the permitted-use clause. |
| Maximum acceptable confidentiality term | Select | No | 2, 3, 5, 7 years, or perpetual OK. Anything longer triggers a redline. Defaults to 3 years when blank. |
| Preferred governing law | Select | No | No preference, or a US state. Governing-law mismatch is the most common NDA redline trigger; if set, any other governing law is flagged. |
| Residuals clause tolerance | Select | No | Never acceptable (strict) · acceptable if narrowed to memory only · fine as drafted. Tech and SaaS counsel split sharply here. Defaults to "never acceptable". |
| Deal / relationship value | Select | No | Under $100K, $100K-$1M, over $1M, strategic. Calibrates how hard to push on liquidated damages, injunctive relief, and indemnification. |
| Require explicit "no LLM training" carve-out? | Yes/No | No | When Yes, flags any NDA that omits an explicit prohibition on using disclosed information to train ML models. |
Use **Try with example data** on the launcher to see a realistic configuration, or point auto-fill at one of the NDAs and let it infer the side, purpose, and governing law.
## How it runs
This is a matrix-review pipeline.
Builds an extraction matrix over the uploaded NDAs. Rows are documents, columns are the clause questions the rubric needs answered.
Schedules per-cell extraction. Each cell is answered against that one document only, with the supporting quote captured.
Blocks until every cell resolves. This is the long pole on large uploads.
Scores each NDA against the benchmark set, calibrated by your side, industry, term ceiling, and residuals tolerance. Produces the per-document risk tier.
Finds cross-portfolio patterns and outliers, then runs the synthesis verifier to drop findings that cite documents not in the set, assert clauses that came back not-found everywhere, or are placeholder filler.
Writes both DOCX artifacts.
## What you get
| Artifact | Format | Contents |
| ----------------------- | ------ | ------------------------------------------------------------------- |
| **Executive Summary** | DOCX | Portfolio-level summary with a traffic-light grid across every NDA. |
| **Per-Document Detail** | DOCX | Per-document benchmark results plus suggested redlines. |
## What this will not do
* Replace counsel review for the final redline decision on a signed NDA.
* Negotiate with the counterparty or send the redline back on your behalf.
* Track changes against a prior version of the same NDA. Use [Document Comparison](/docs/guides/document-comparison) for that.
* Investigate facts outside the uploaded NDA, including counterparty background or related deal terms.
* Confirm enforceability of trade-secret claims under specific state Uniform Trade Secrets Act variants.
## Related
The same shape for master service and SaaS agreements.
Score against your own playbook rather than market benchmarks.