# Try an operations review with a workbook, brief, and task handoff.

Reviewed: 2026-10-25

Canonical: https://clankercloud.ai/templates/operations-review-pack

Use this fictional operations packet to evaluate a Workspace Agent across data analysis, Office deliverables, and task handoff. Start with the brief and operating data, check a workbook and management summary, then supply a dated correction. The expected results below are authored reviewer checks; they are not observed Agent outputs or a customer case study.

## What you are working toward

Your own recorded assessment of whether the saved files and handoff agree with the supplied evidence.

## Start with

- A new Project for the fictional Northstar Service Desk packet

- The brief and operating data; keep the correction for the second pass

- A reviewer who checks the authored expectations

## Download the first two files and keep the correction separate

Add the brief and operating data to a new Project. Ask for a workbook and a one-page management brief for 19–23 October 2026. Keep the later correction out of the first request so you can test a meaningful revision.

Ask for the requested native file types and inspect them. Add a short slide presentation only after the workbook and brief agree. These are deliberately small authored inputs; passing this trial does not establish performance on your own larger or messier data.

## Reviewer reference for the first pass

The packet has five daily records. Blank refund data on Wednesday means the refund total for the full period is incomplete. The brief identifies an approved reporting period and a proposed staffing change; a proposal is not a commitment.

| Check | Expected from the supplied files |
| --- | --- |
| Tickets received | 500 across five records. |
| Tickets resolved | 470 across five records. |
| Opening and closing backlog | 40 opening; 70 calculated closing: 40 + 500 − 470. |
| Refunds | Known recorded refunds total $300; Wednesday is unknown, so the full-period total is incomplete. |
| Staffing | One extra Friday shift is proposed; approval and owner are unknown. |

## Supply the second-pass correction

The 24 October correction confirms Wednesday refunds of $40 and approves the extra Friday shift with Rowan as owner for the next review cycle. The corrected full-period refund total is $340. Ticket counts and closing backlog stay unchanged.

Ask the Agent to revise the workbook, management brief, and any deck. Check that each file uses the corrected refund total and approval state. Attach the reviewed versions and a concise note to a Planner task if you are testing the handoff. Creating the task must not be treated as starting another run.

## Record what you actually observed

For each check, record pass, revise, or not observed with the output filename and version. Include manual corrections and whether the actual Office files opened and rendered. A confident explanation without the saved artifact is not a pass for file creation.

If the method is useful, turn its source-checking and revision rules into a custom Agent. Keep these fictional records out of its permanent business knowledge; they are test data. Try a different packet before concluding the method generalizes.

## Sample files

Fictional inputs. Download and add to your Project.

- [operations-brief.md](https://clankercloud.ai/templates/operations-review-pack-files/operations-brief.md): Fictional audience, source rules, and initial approval state.

- [operating-data.csv](https://clankercloud.ai/templates/operations-review-pack-files/operating-data.csv): Five fictional daily records, including one unknown refund value.

- [dated-correction.md](https://clankercloud.ai/templates/operations-review-pack-files/dated-correction.md): Second-pass input: confirmed refund and an approved staffing decision.

## Try this prompt

Copy into a Workspace Session. Opening Workspace does not send it automatically.

```text
Use operations-brief.md and operating-data.csv in this Project to save an XLSX analysis and a DOCX management brief for 19–23 October 2026. Inspect the data structure first. Preserve missing values, distinguish proposals from approved decisions, and show how the closing backlog is calculated. Save and inspect the actual files. Tell me what remains unknown. Do not use a later correction until I supply it, and do not send or publish anything.
```

## Illustrative result

Fictional sample; your result depends on your inputs and needs review.

```text
Authored reviewer reference, not an observed Agent output: initial tickets received 500; resolved 470; closing backlog 70. Known refunds $300 with Wednesday unknown. After the correction: complete-period refunds $340; extra Friday shift approved with Rowan as owner. Ticket totals and backlog remain unchanged.
```

## Before using the result

- First pass: 500 received, 470 resolved, closing backlog 70.

- First pass: known refunds $300 with Wednesday explicitly unknown.

- First pass: extra shift remains proposed; approval and owner remain unknown.

- Second pass: full-period refunds $340 and approved shift owned by Rowan in every dependent output.

- Saved files open, layout was inspected, and the Planner handoff identifies reviewed versions.

## Common questions

### Are these results from a real Agent run?

No. The files are fictional inputs and the expected results are authored reviewer checks. Run the packet yourself and record the files and behavior you actually observe.

### Can I use this packet to test a custom Agent?

Yes. Keep the same acceptance criteria across candidates, withhold the second-pass correction initially, and compare saved outputs. Use a different packet before reusing the Agent for consequential work.

## Sources

- [Office workflow](https://clankercloud.ai/learn/create-office-deliverables-with-ai)

- [Planner handoff](https://clankercloud.ai/learn/ai-planner-tasks-and-deliverables)

- [Agent acceptance testing](https://clankercloud.ai/learn/test-ai-agent-before-reuse)

## Related

- [What can you do in Clanker Cloud Workspace?](https://clankercloud.ai/learn/what-can-clanker-workspace-do)

- [Turn Project files into Office deliverables you can review.](https://clankercloud.ai/learn/create-office-deliverables-with-ai)

- [Keep the task, progress, and deliverable together in Planner.](https://clankercloud.ai/learn/ai-planner-tasks-and-deliverables)

- [Give the Agent a test with an answer you can check.](https://clankercloud.ai/templates/agent-acceptance-test)

- [Build AI Agents for the work your business repeats.](https://clankercloud.ai/learn/ai-agent-factory)

[Open Workspace](https://workspace.clankercloud.ai)

[Current pricing and usage](https://clankercloud.ai/pricing)