Documentation For agents

Use the command-line tool

The sendbop command-line tool calls the SendBop API with an agent's or your own token. It reads the list of operations from the API description, so every API operation is available with the same permissions as the workspace.

Get it

The tool is not available to download yet. Until it is, call the HTTP API directly: it offers the same operations with the same token.

Configure it

Set two environment variables in the agent's environment or secret store:

export SENDBOP_URL=https://sendbop.example
export SENDBOP_TOKEN=<the token from Agents>

SENDBOP_URL is the address of your SendBop workspace, without a path.

Confirm the connection

  1. sendbop operations lists every operation.
  2. sendbop get-discovery --mspId <id> returns your permissions, limits and the released capabilities. Your account ID is shown under Team & sharing, Sharing.
  3. sendbop help prepare-plan shows an operation's flags and body fields.

Everyday commands

  1. sendbop get-portfolio --mspId <id> lists work waiting on someone.
  2. sendbop get-case-packet --mspId <id> --caseId <id> returns what an investigation knows.
  3. sendbop prepare-plan --mspId <id> --domainId <id> --body '{"kind":"dmarc","purpose":"..."}' prepares a change; nothing is published.

Output and errors

Output is JSON. Exit codes: 0 success, 2 usage, 3 authentication or permission, 4 not found, 5 conflict or stale state, 6 limit or outage. Lists return one page and say whether more exist.

Rotating the token

Rotate from Agents; the old token stops working at once. Update SENDBOP_TOKEN wherever the agent runs.

Documentation version docs-2026-10-12.

Report an error or ask a question