Flags the clause. Keeps the paper trail on who signed off.
Reads incoming contracts against a company's playbook, redlines what deviates, and routes anything ambiguous to the right person — logging exactly who reviewed what, and when.
Reads incoming contracts against a company's own playbook, not a generic legal-AI baseline.
Proposes specific redlines for clauses that clearly deviate from playbook standards.
Sends anything unclear to the right person, and logs exactly who reviewed what, and when.
Planned design — this pipeline does not exist yet.
Reads an incoming contract's clauses against the company's own stored playbook standards.
Proposes specific redlines for clauses that clearly fall outside playbook standards, citing the relevant playbook section.
Sends non-standard or unclear clauses to the appropriate person, recording who reviewed what and the decision made.
Illustrative — no such log exists yet; this is what a completed review trail is designed to look like.
MSA_Draft_v3.docx — compared clause-by-clause against playbook
Liability cap floor unstated — deviates from Playbook §4.2
Sent to legal for a decision on the flagged clause
Reviewer, timestamp, and outcome recorded to the trail
The reasoning approach this concept is designed around — illustrated here as the comparison behind one flagged clause.
Liability cap: total fees paid in the twelve (12) months preceding the claim. Cap floor: none below 12-month fees.
Liability cap: total fees paid in the twelve (12) months preceding the claim. Cap floor unstated.
Illustrative only — the actual comparison logic and what counts as "clear-cut" are unspecified.
Grounds every redline in a specific company's own playbook section, not a general notion of 'standard' contract terms.
Auto-redlines only where the deviation is clear-cut; routes anything non-standard to a person rather than guessing.
Tagline leads with 'keeps the paper trail' — who reviewed what and when is treated as core value, not a side effect.
Like MirrorSense, ThreatSense, and ChainSense, this is new build work with no Decision Intelligence or Data Reliability foundation.
Nothing below is built. This describes the intended design and its dependencies — sketched as the document-review screen ClauseSense would show a reviewer.
7.1 Term. This Agreement shall commence on the Effective Date and continue for a period of thirty-six (36) months, unless earlier terminated in accordance with Section 11.
7.2 Limitation of Liability. Except for breaches of Section 9 (Confidentiality), in no event shall either party's aggregate liability arising out of this Agreement exceed the total fees paid by Customer in the twelve (12) months preceding the claim.
9.1 Confidentiality. Each party shall protect the other's Confidential Information using no less than a reasonable standard of care...
Liability cap language present, but the cap floor is undefined in this concept.
Every redline traces to a specific section of the company's own playbook.
Ambiguous clauses go to a person, not a forced auto-redline.
Logs exactly who reviewed each clause and what they decided.