Skip to main content
Human guide: work with your agent → Convert one call record into proposed CRM changes. Link each proposed field value to the supporting call evidence.

​
Before you start

Use an authorized workspace and a small file with stable source IDs. Have source access and your team’s acceptance policy available. This guide describes customer-authored logic, not an automatically enabled product policy.

​
1. Understand the decision

A meeting-ended event does not necessarily mean the transcript is available. Choose a supported readiness signal or explicit retrieval check. Test this fictional sample before calling providers:

​
2. Inspect supported operations

Inspect a returned ID with deepline describe <returned-id> --json. Compare required identifiers, outputs, access mode, and price. A similar title does not establish that a contract fits.

​
3. Implement the bounded workflow

Ask your agent to author post-call-crm-automation.play.ts using Write Plays. Have it declare the CSV input and preserve source IDs. Reuse existing contracts when they fit the transformation. Implement these rules explicitly:
  1. Distinguish meeting completion, transcript readiness, and retrieval failure.
  2. Retrieve the authorized transcript and resolve participants to CRM identities.
  3. Preserve the supporting passage for each proposed field or task.
  4. Keep unstated dates and owners unknown. Reconcile uncertain writes before another attempt.
Keep external changes disabled for the first pilot. That control belongs in the authored logic, not a universal Play dry-run flag.

​
4. Check and pilot

Prepare pilot.csv using CSV-safe selection. Then check the source:
Resolve authoring errors first. After reviewing source and price, run only the pilot:
Use the ID returned by the pilot. Select a returned dataset path when several exist. Provider-backed execution can incur usage.

​
5. Verify the outcome

Trace each proposal to the transcript passage. Keep source retrieval failures separate from meetings with no agreed next step.
Every source ID has a result or explicit exception. The fictional cases follow the stated policy, and actual results retain their evidence.
Ask your agent why a record was accepted, held, or rejected. Correct the rule before expanding input or enabling external writes.

​
Continue

Follow-up drafting covers commitments. Monitor setup applies only to supported feeds.