Update Trigger
update_triggerA send step's message, or a connection request's note, lives in its message_templates, so
to change what a step sends,
edit the template with update_node; an on_enter prompt on that step carries only
extra instructions (tone, whether it may rephrase, what to write when a slot can't be
filled), never the message itself.
You MUST call get_skill_guide('trigger_code') before writing code — the
qualification rules, sandbox globals, tool surface, and out contract live
in the skill. New or changed code runs on the trigger's next firing.
The agent's goal is injected into every turn on this agent, so it steers later
runs and replies rather than merely describing them: a trigger edit that
changes what the agent does leaves the goal asserting the old behavior. After
this call, check the goal — when it describes what you just changed (a step
you rewrote, a daily volume or cadence it states, a trigger it says is
paused), bring it in line via update_agent(agent_id=..., goal=...) in the
same turn.
Dict with success, agent_id, trigger_id, next_run_at, and the edited trigger's
full dict under trigger, plus a warning when an active trigger sits on a paused
agent, since it doesn't fire until the agent is set back to active.
Read the agent's other triggers via get_agent_details.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| code | No | New trigger code. Runs on the trigger's next firing. | |
| label | No | New label. | |
| prompt | No | New per-run instructions for this trigger. Name any outside resource a run reads (a Google Sheet, doc, file, URL) by the identifier its tool takes (the spreadsheet ID and tab, the URL), not only by its title — a later run doesn't see this chat. | |
| status | No | 'active' to resume firing, 'paused' to stop this trigger. | |
| trigger_id | Yes | ID of the trigger to edit (from the agent's trigger list). |