Onboarding · ← Use cases

What the team has actually decided before the new joiner shows up

The working agreements, the unwritten rules, the artefacts - ready for the inbox.

Best for
Pre-departure / pre-arrival hand-off, week before someone joins or moves
Primitives
DecisionsInteractionsStance
Situation
Pre-departure

Connect Paperbase. Plug your agent into Paperbase over the API today - MCP support is coming soon. To get a key, sign in and open Settings → API keys, then hand it to any agent that speaks tools - Claude, Cursor, ChatGPT, your own. Paste a recipe, and you are running.

Side by side

~60% fewer tokens on this recipe

Without Paperbase

~$210k tokens
  1. Ask the team for their running notes.
  2. Open the most recent standup thread.
  3. Compose by hand.

With Paperbase

~$60k tokens
  1. For Team {{team}}, pull every Decision in the last 90 days whose audience included this team.
  2. Filter to Decisions tagged Consequence=Working-Agreement, Style-Guide, or Process.
  3. For each, pull the originating meeting or doc and any artifact attached (links).
  4. Compose a one-page briefing: the Decisions (one line each), the artefacts (one line each), the agreed Stance per major topic.

Agent prompt

You're prepping the working-agreement hand-off for {{new_joiner}}. {{team_lead}} wants one page the new joiner can read in the inbox on day one.

Using Paperbase:

1. For {{team}}, pull every Decision in the last 90 days whose audience included this team.
2. Filter to Decisions tagged working-agreement, style-guide, or process.
3. For each, pull the originating meeting or doc and any artifact attached (links).
4. Compose a one-page briefing: the Decisions (one line each), the artefacts (one line each), the agreed Stance per major topic.

Output:

- One-page briefing: the Decisions, the artefacts, and the Stance per major topic. Citations to source.
Placeholders in {{double_braces}} are inputs the agent will ask for at runtime - new joiner, team lead, team. 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.