‹ Executive recipesExecutive & Chief of Staff · Agent Blueprint

Monthly strategy review

Strategic initiatives that lost momentum this month. Where the company stopped talking about something, and what that means.

Best forMonthly leadership review, strategy cadence
PrimitivesInitiativesDriftTrends

Why this is memory-only: a persistent memory layer preserves the longitudinal pattern that ordinary search cannot see.

Side by side: token usage, with and without Paperbase

Memory only

Without Paperbase

Not recoverable from search alone

  1. Manually list active initiatives from memory.
  2. Skim recent meeting notes and Slack for each one.
  3. Try to recall when each was last seriously discussed.
  4. Guess whether silence means done, parked, or forgotten.
  5. Miss the initiatives nobody happens to mention this week.

With Paperbase

Persistent memory layer

  1. Ask Paperbase for each initiative's mention history over the quarter.
  2. Let it flag where the cadence collapsed: from weekly to nothing.
  3. Pull the last interactions before the silence, so the reason is visible.
  4. Compare against commitments made to each initiative.

Agent prompt

You are running the monthly strategy review.

Using Paperbase:

1. List the active strategic Initiatives and their owners.
2. For each, return the mention history over the last 90 days - meetings, threads, decisions.
3. Flag initiatives whose mention cadence collapsed: the last time it was seriously discussed, and what was said then.
4. Note any commitments attached to each initiative and their status.

Output:

- Initiatives that lost momentum this month, ranked by how abrupt the silence was.
- For each: last discussion (quote + date), what was committed, and what the silence likely means.
- A one-line question for the leadership team on each flagged initiative.

Placeholders in {double_braces} are inputs the agent will ask for at runtime. Give the prompt to any agent connected to Paperbase, and the rest grounds in your own memory graph.

Want a deeper recipe for your team?

Send us the prompt, the source systems, and the workflow. We will draft the recipe and sign you into the sandbox to run it.