What does a triage agent that runs every 10 minutes cost, and what limit do I need to raise?

A Tares triage agent that runs every 10 minutes costs well under a cent a run, but you need to raise the default limit of 50 runs per agent per day. A 10 minute cadence is 144 runs a day, so raise the limit under Settings → Agents or with TARES_AGENT_DAILY_CAP, and give the agent a budget. The main agent runs only when triage flags something, at most once per cooldown per entity.

How it works

The pattern in Watch, then escalate uses two agents linked by a handoff. The triage agent is cheap and frequent. The main agent is thorough and rare.

• The schedule trigger fires every interval for all the events of its sources. At every: 10m that is 144 runs a day.
• The summary keeps each run small. The trigger hands the triage agent a summary of the window instead of raw lines: for each label you name, the count per value in this window against the window before, largest change first, plus a few recent lines. A stream of tens of thousands of lines becomes a page the model can read, which is why a run is cheap.
• The default limit is 50 runs per agent per day. A 10 minute cadence is above it, so the triage agent hits the cap partway through the day unless you raise it.
• The main agent is gated. It has no trigger of its own. It runs only when triage concludes with the verdict investigate, and the handoff cooldown limits it to at most once per cooldown per entity. With a cooldown of 30m, that is at most two root-cause runs per hour per entity.

Most triage runs end as no_op: the run succeeds with no finding and a one-line reason. Those runs are the cheap ones, and they are most of the 144.

Set it up

1. Add the schedule trigger. On the project’s page click + next to Triggers, then fires: on a schedule. Or in the catalog:

triggers:
  - name: watch-web
    sources: [web_logs]
    key_field: service
    condition:
      every: 10m                          # at least 1m
      summary_by: [service, status_code]  # labels to count; default: the trigger's key field

2. Raise the daily limit. Open Settings → Agents and set the runs per agent per day above 144. You can also set the limit with the TARES_AGENT_DAILY_CAP environment variable. Leave some headroom, because the default of 50 is below what a 10 minute cadence needs.
3. Create the triage agent on watch-web. Seed its prompt from the Triage (on a schedule) preset. Pick the cheapest model on your provider that calls tools reliably.
4. Give the agent a budget. The budget is what keeps the cadence from running up a bill. Set it when you create the agent.
5. Link the main agent with a handoff. Create the main agent without a trigger, seed it from Root cause after a handoff, then add a handoff on the triage agent under When it concludes. In the catalog the handoff looks like this:

agents:
  - name: triage
    trigger: watch-web
    concludes: true
    verdicts:
      - { verdict: investigate, when: one entity looks like a problem and needs a closer look }
    prompt: ...             # from the Triage (on a schedule) preset
    handoffs:
      - { verdict: investigate, agent: rca, cooldown: 30m }   # at most two root-cause runs per hour per entity
  - name: rca               # no trigger: it runs only when handed off
    concludes: true
    verdicts:
      - { verdict: rca, when: you found what is failing and why }
      - { verdict: resolved, when: it was noise, or it is already over }
    prompt: ...             # from the Root cause after a handoff preset

6. Check the spend. Open the triage agent’s runs under Agents. Each row shows its outcome, its reason and its cost, so you can see what a run costs on your provider. Use one link between the agents, either a handoff or a trigger over findings, never both, or the main agent runs twice per flag.

What you get

• A predictable triage cost. 144 runs a day, each a small model reading a summary, each well under a cent.
• A hard ceiling. The raised daily limit and the agent’s budget cap what the cadence can spend.
• Rare, thorough root-cause runs. The main agent starts only when triage flags an entity, and the cooldown caps repeats on that entity.
• Cost on every run. The runs table shows mostly “no finding” rows, each with its reason and cost. A flagged run shows an investigate chip with its headline.
• A full record. The tick, the triage verdict, the handoff and the root-cause note all sit on the entity’s timeline.

If you want to stop spending, disable the triage agent or pause its schedule trigger. With no new investigate findings, the main agent goes quiet too.

When something else fits

• If you want an agent to wake only when something specific happens, an alert-driven agent fits better. See Build an AI SRE for that pattern, which this one sits in front of.
• If a 10 minute check is more often than you need, lengthen every. The trigger accepts any interval of at least 1m, and a longer interval means fewer runs a day and a lower cost.

Related