Docs › Pydantic AI and smolagents scheduled run monitoring with RunVouch
Pydantic AI and smolagents scheduled run monitoring with RunVouch
Pydantic AI and smolagents scheduled run monitoring at the script level: these libraries are a few hundred lines you call from a script. The script is what runs on cron, so the script is what gets wrapped; no callbacks are required for the basic detectors.
How it runs on Pydantic AI and smolagents
Wrap the script with rv run for MISSED, FAILED, STALLED, DRIFT and evidence. Add run.tool() calls from a tool wrapper if you want retry-storm and per-call cost.
Store the key
An environment variable in the cron environment.
Wrap the job
# crontab
0 6 * * * rv run morning-digest --evidence-file /out/digest.md -- python digest.py
# digest.py (Pydantic AI); smolagents is the same shape
from pydantic_ai import Agent
agent = Agent("anthropic:claude-sonnet-5", system_prompt="...")
result = agent.run_sync("Summarise overnight changes")
open("/out/digest.md", "w").write(result.output)rv fails open: if RunVouch is unreachable the job still runs and you get one warning line.
Register the cadence and caps
rv agent nightly-report --cadence 24h --grace 30m --max-runtime 1h --evidence --cap-run-cost 2
Register the agent once, from anywhere with the key. Cadence is what turns a schedule that stopped into an alert; --evidence makes a run without evidence a failure; the caps pause the agent when it overspends.
What goes silent on Pydantic AI and smolagents
- A script that raises exits non-zero and is FAILED with the stderr excerpt in the alert.
- A script that writes an empty digest is NO_EVIDENCE.
- Output size drifting from 6 KB to 300 bytes over a week is DRIFT, which is how a broken data source usually shows up first.
What Pydantic AI and smolagents does not tell you
Both libraries return a result object with usage counts and raise on hard errors; Pydantic AI can send spans to Logfire if you configure it. Neither has a schedule, a budget, or a notion of whether the output was written anywhere. What runs them is cron, and cron's only failure channel is local mail that nobody reads.
What RunVouch detects
| Alert | What it means here |
|---|---|
| MISSED | cadence plus grace passed and no run started |
| FAILED | non-zero exit or a reported failure, with the stderr excerpt |
| NO_EVIDENCE | the run said ok but the file, URL or assertion you required is missing |
| STALLED | a run started and never ended within max runtime |
| RETRY_STORM | the same tool called with identical input many times in one run |
| BUDGET_RUN / BUDGET_DAY | cost cap crossed; the agent is paused until you resume it |
| DRIFT | duration or output size far off its 7-run baseline |
Set up in two minutes
- Get a free key (3 agents, no card) and store it where this page says.
- wrap the job: copy the snippet above into the scheduled job.
- Register the cadence once:
rv agent nightly-report --cadence 24h --grace 30m --evidence, or let the first run create the agent and set the cadence on the dashboard. - Send one test alert:
curl -X POST https://api.runvouch.com/v1/settings/test-alert -H "X-API-Key: $RUNVOUCH_KEY". The next missed, failed or empty run reaches the same channels.
Also running agent frameworks?
Dify and Flowise. Ollama and local LLM batch jobs. OpenClaw (verified-run skill). LangGraph. The full list of runtimes is on integrations, and the same agent can report from several of them: one job, one cadence, wherever it happens to run.
Need a key? Get a free key · Stuck? contact