PLAYBOOK DRIVEN
Security and compliance
Contract review agent
Clausewise
Reviews an agreement against your own playbook, marks what departs from it, and drafts the redline you would have written.
200 tokens/s
Throughput
98.7%
Accuracy
820ms
Median latency
4.6M
Calls per month
Extracts obligations, compares each against the firm's positions, and ranks deviations by exposure rather than by order of appearance.
Scope
Extracts obligations, compares each against the firm's positions, and ranks deviations by exposure rather than by order of appearance.
Capabilities
- Playbook comparison
- Obligation extraction
- Deviation ranking by exposure
- Redline drafting
- Prior-version diffing
Runs on
Deployed in
Calling it from the SDK
const counsel = new NexusAgent({ role: 'contract_review', playbook: 'legal://positions-2026' });
const review = await counsel.review({
document: agreement,
rank: 'by_exposure'
});
Recorded exchanges
Published examples, reproduced verbatim. This page does not call a model.
Where this MSA departs from our positions
Asked
Review this master services agreement against our playbook and rank what matters.
Returned
Eleven departures from the playbook. Three are worth partner time; the rest are drafting preferences. 1. Liability cap (§11.2) — uncapped for data breach, against our position of 3× fees. On this contract value that is a difference of roughly two orders of magnitude in worst-case exposure. Redline drafted. 2. Governing law (§18) — their jurisdiction, and the arbitration seat differs from the governing law. That combination is workable but expensive; the playbook prefers them aligned. Redline drafted with a fallback. 3. Termination for convenience (§14.1) — available to them at 30 days, to us at 180. Asymmetric in a way the playbook flags specifically. The remaining eight are noted in the margin with suggested language. This is a first-pass review by a machine and is not legal advice — it needs a qualified reviewer before it goes back to the counterparty.