Back to use cases
Finance, Procurement, Legal

Verbal commitments that conflict with the contract draft

What sales promised on the call vs. what made it into the paper: flagged before signature.

Best for
Contract close
Primitives
CommitmentsContractsDiff
Token savings
60%

The prompt

You are reviewing the {{account_name}} contract before signature. Using Sentra: 1. Pull every verbal Commitment Sentra extracted from meetings, threads, or emails between our team and {{account_name}} during the deal cycle. 2. Pull the current contract draft (Ironclad / Docusign / shared doc). 3. Diff the two: which verbal Commitments are reflected in the written terms, which are missing, and which written terms contradict what was said. 4. Categorize each gap by risk: high (legal exposure, financial commitment, deliverable date), medium (process / scope clarification), low (nice-to-have). Output: - A diff table: verbal commitment · source · status in contract · risk. - A short paragraph: the one or two clauses that should be raised with legal before signature. - Sentra citations for every verbal commitment so the legal team can verify the exact words.

See it work

Run it from

ClaudeCursorChatGPTAny agent · MCP

How it runs

  1. 01Copy the prompt into any agent connected to Sentra over MCP.
  2. 02The agent reads the company memory, scoped to what you are allowed to see.
  3. 03One grounded answer comes back, with every line traceable to its source.

Or automate it with Actions

The same prompt runs as a Sentra Action: on a schedule, or on a semantic trigger. Output lands in Slack or email, and every send waits for your approval.

Trigger idea: contract close

Sentraover MCPVerbal commitments checked against the paper

Solara · the calls vs. the draft

Credit rollover

call · Aug 28 · 00:19:05

Naomi said credits roll over twelve months on the Aug 28 call; the draft says six.

Audit logs

call · Sep 2 · 00:07:44

"Included in your tier" on the Sep 2 call; the draft lists audit logs as a paid add-on.

Uptime

review · Aug 21

99.95 promised verbally in the security review; the draft carries 99.9.

The prompt above becomes this. Fictional data, real mechanics: every line cites the meeting, thread, or record it came from.

Without Sentra

  1. 01Pull every call transcript, email, and Slack thread between our team and the customer during the deal cycle.
  2. 02Read each to identify verbal promises — discount terms, deliverable dates, exception clauses.
  3. 03Pull the current contract draft from Ironclad or Docusign.
  4. 04Manually diff each verbal promise against the written terms — present, missing, or contradicted?
  5. 05Categorize each gap by risk: high (legal / financial exposure), medium (scope clarification), low (nice-to-have).
  6. 06Compose the diff table with citations to verbal moments.

~95,000 tokens

With Sentra

  1. 01Pull all verbal `Commitments` extracted from `Interactions` between our team and the account during the deal cycle.
  2. 02Pull the current contract draft.
  3. 03Diff each verbal Commitment against contract clauses — already segmented by clause type.
  4. 04Risk-rank each gap.
  5. 05Compose the diff table with Sentra citations.

~38,000 tokens

Book a DemoStart Free