# Example: source-linked decision memory

> A source-linked decision memory turns a work artifact into a concise, auditable record an agent can reuse.

## Source to memory

Source artifact: Granola call.

Source excerpt: The onboarding sweep asks which app sources are included before the agent scans anything.

Ambience memory:

- Type: decision
- Title: Ask before including sources in onboarding sweeps
- Body: Onboarding sweeps must show the available app and local sources, then ask the user which ones are included before proposing Ambience memories.
- Scope: org
- Tags: onboarding, source-linked

Controls:

- Source: Granola call remains the source of record.
- Redaction: customer names removed before storage.
- Audit: accepted by product owner during onboarding review.

Future reuse:

- A future onboarding task retrieves the decision before changing the sweep flow.
- The UI can show the linked source without copying the whole call into Ambience.

## Why Ambience

- The source stays visible.
- The decision stays concise.
- Future agents can retrieve the memory before changing onboarding.

## Related

- [Granola call to memory](https://ambience.sh/examples/granola-call-to-memory)
- [How to audit AI agent context](https://ambience.sh/answers/how-to-audit-ai-agent-context)