Back to use cases
Sales

Open opportunities with overdue commitments from our side

Which deals are stalling because we owe the customer something we never delivered.

Best for
Weekly pipeline review
Primitives
OpportunitiesCommitmentsOwners
Token savings
72%

The prompt

You are auditing the pipeline for self-inflicted slippage. Using Sentra: 1. List every open Opportunity (any stage past discovery). 2. For each, join Commitments where direction is "from us, to this account" and status is not yet "delivered". 3. Filter to commitments whose due date is in the past, or whose extracted urgency has lapsed. 4. For each, identify the original Interaction the commitment came from and the named owner on our side. Output a single table, sorted by days overdue, descending: | Account | Opportunity | Commitment | Owner | Due | Source | Then in two sentences: which commitments, if closed this week, would unblock the most ARR.

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: weekly pipeline review

Sentraover MCPOpen opportunities · commitments overdue from our side

3 deals stalling on our own promises

Solara · $180k renewal

call · Sep 2

Revised order form owed by Naomi, three days late. Procurement will not schedule the next call without it.

Meridian · $220k renewal

email · Aug 30

Security questionnaire owed by Priya, six days late. Their lead marked the deal paused-waiting in the last thread.

Northgate · $75k new

thread · Aug 26

Pilot success criteria doc promised by Alex, never sent. The prospect has asked for it twice.

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 open opportunity from the CRM. Parse stage, dates, and raw notes fields.
  2. 02For each opportunity, search call transcripts and emails for promise-shaped language from our team to the customer.
  3. 03Manually classify each as a real commitment vs. casual language. Determine direction (from us vs. to us).
  4. 04Manually compute due dates from natural-language references ('by end of month', 'before the quarter ends').
  5. 05Cross-check against any internal tracker. Resolve discrepancies between what was said and what was logged.
  6. 06Identify the owner on our side for each commitment by reading the source thread.
  7. 07Compose the table. Sort by days overdue.

~150,000 tokens

With Sentra

  1. 01Pull every open `Opportunity` with stage past discovery.
  2. 02Join `Commitments` where direction = from-us-to-account and status ≠ delivered.
  3. 03Filter on the typed `due_at` field — past-due commitments are already structured.
  4. 04Each Commitment carries its originating `Interaction` source and the named `Actor` owner.
  5. 05Sort by days overdue. Compose the table.

~42,000 tokens

Book a DemoStart Free