Back to use cases
Product

Where are two teams making conflicting decisions?

Two Decisions, overlapping scope, opposite polarity: caught before they ship.

Best for
Cross-team architecture reviews
Primitives
DecisionsTeamsPolarity
Token savings
75%

The prompt

You are scanning for cross-team decision conflicts. Using Sentra: 1. Pull every Decision made in the last 60 days, with Sentra's extracted scope (which Value Objects it touches), the originating team, and the polarity. 2. For each pair of Decisions whose scopes overlap, evaluate whether the polarities or implementations are mutually contradictory. 3. Surface only the conflicts where both Decisions are still live (neither has been superseded). Output: - A list of up to ten live conflicts, each: Decision A (team, date, statement), Decision B (team, date, statement), the overlap, and the practical contradiction. - For each, the named Actor on each side who should be looped into a resolution conversation, and a one-line suggested framing.

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: cross-team architecture reviews

Sentraover MCPTwo teams, overlapping scope, opposite polarity

One collision found

Platform · Aug 12

platform sync · Aug 12

Decided to deprecate the v1 webhook API in Q4 as part of the event-bus consolidation.

Solutions · Aug 19

workshop · Aug 19 · 00:52:40

Promised Harbor continued v1 webhook support "through mid-2027" during the integration workshop.

Why nobody caught it

Neither team was in the other's room, and both decisions are individually reasonable. Suggested owner for the reconciliation: Alex.

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 meeting transcript and Slack thread from the last 60 days across all engineering and product teams.
  2. 02Read each to identify decision moments. Manually extract: team, scope (which systems touched), and polarity (for/against, build/skip).
  3. 03Cross-join every decision pair. For each pair, judge whether scopes overlap by re-reading both.
  4. 04For overlapping pairs, judge whether the polarities are mutually contradictory.
  5. 05Check whether each decision is still live — has anyone reversed it? Re-search for follow-ups.
  6. 06Identify the named owner on each side for resolution.
  7. 07Compose the list of live conflicts.

~190,000 tokens

With Sentra

  1. 01Pull all `Decisions` from the last 60 days with extracted `scope` (Value Objects touched), team, and `polarity`.
  2. 02Cross-join: pairs where scopes overlap and polarities contradict.
  3. 03Filter to pairs where both are still live (no superseding Decision).
  4. 04Pull the named `Actor` on each side from Sentra's owner relation.
  5. 05Compose the conflicts list with suggested framing.

~48,000 tokens

Book a DemoStart Free