Back to use cases
Engineering & SRE

Sprint blockers and how they were resolved

Every blocker mentioned in standup or thread, and the resolution path.

Best for
Retro prep
Primitives
BlockersInteractionsOwners
Token savings
63%

The prompt

You are preparing a retro on sprint {{sprint_id}}. Using Sentra: 1. Pull entities around the sprint window where Sentra extracted a Blocker — standups, threads, 1:1s, ticket comments. 2. For each Blocker, return: who raised it, when, the affected Value Object (service / feature / project), the resolution path, the owner who unblocked it, the time-to-unblock, and whether it shipped resolved or rolled to next sprint. 3. Cluster recurring root causes (dependency wait, missing context, unclear scope, broken tooling, etc.). Output: - Table of every Blocker with the columns above. - Top three recurring root causes with the cluster size. - One paragraph: the single change that would have prevented the most pain this sprint.

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: retro prep

Sentraover MCPEverything called a blocker, and what happened next

Sprint 42 · blockers and resolutions

Staging data refresh · resolved

standup · 2 mentions

Broke twice mid-sprint; Priya pinned the snapshot and documented the pin in the runbook.

Payments sandbox creds · resolved

thread · Aug 27

Expired mid-integration. Two days lost to the vendor ticket; renewal is now calendared.

Design tokens PR · open

standup · Sep 2

Stuck four days awaiting review; the named reviewer is out and no fallback was assigned.

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 standup transcript, threads in #eng channels, 1:1 notes, and ticket comments in the sprint window.
  2. 02Read each for blocker-shaped language ('blocked on', 'waiting for', 'can't ship until').
  3. 03For each blocker, identify the affected service or project by reading context.
  4. 04Track resolution path — search subsequent threads for the unblocking moment.
  5. 05Compute time-to-unblock manually from timestamps.
  6. 06Cluster recurring root causes by re-reading and judging.
  7. 07Compose the table and root-cause ranking.

~120,000 tokens

With Sentra

  1. 01Pull `Interactions` in the sprint window where Sentra extracted a `Blocker` event.
  2. 02Each Blocker carries: typed `Actor` raiser, affected `Value Object`, resolution path, owning `Actor`, time-to-unblock, shipped or rolled.
  3. 03Cluster recurring root causes via Sentra's learned `Theme` taxonomy.
  4. 04Compose the table and the closing recommendation.

~44,000 tokens

Book a DemoStart Free