Sume schedule cap null or 0: what a per-run override can do
A per-run generation_spend_cap_usd lowers a Sume schedule's cap but never raises it; null drops that ceiling, 0 is rejected, and wallet limits still apply.

A scheduled agent run spends money while nobody is watching, so the cap rules matter more than the cron expression. Sume's docs give four rules for the per-run cap.
| You send | Result | Source |
|---|---|---|
| Nothing, schedule cap unset | $1.00 default (1000000 USD micros) | Create a schedule |
| A number above 0 | min(request, schedule cap): lowers, never raises | Scheduled, limits |
| null | Runs without the automation ceiling | Scheduled, limits |
| 0 | Rejected by the API | Scheduled, limits |
What null does and does not mean
null removes the schedule's own ceiling for that run. It does not remove the rest: the docs say wallet balance, generation admission and org limits still apply. So null is not unlimited spend, but it is a larger blast radius than your saved cap, and an unattended caller should have a reason to use it.
A safe pattern
Set the schedule cap to the most one run should ever cost, in the dashboard. Let callers send a smaller generation_spend_cap_usd for cheap runs. Treat null as a code-review item: grep your repo for it. Remember the cap bounds generation, not the agent's own LLM turn, which the Runs and results page says bills a separate wallet.
The cap is read from the schedule object as generation_spend_cap_usd_micros, where null means the $1.00 default applies. When you call the API trigger, send the override in the request body, and add an Idempotency-Key so a retry does not run twice.
Sources
Related posts
More in Agents
- tts_create dry_run will not warn about a voice-language mismatch
A cost-only dry_run on Sume's tts_create stays a preview; the language double-check runs on submission. How to preflight voice, language and cost together.
- tts_source_get and verify_spine: check a voiceover vs its script
Sume's hosted MCP has two free read tools for voiceover work. One returns the accepted script for tts_create, the other compares chosen TTS jobs to it.
- Weekly new-model watcher as a Sume schedule: schema without links
A scheduled watcher that lists new AI models fails if its output schema holds vendor page URLs. Use host-only fields and read each run with the actions API.
- Weekly trend video schedule stops early: the $1.00 default cap
A Sume schedule with no spend cap runs with $1.00 of generation per run, so a weekly video run can stop after a clip. Set the cap to fit the plan.
Written by Sume