run-without-you

1# Phase 4 — Run Without You (the recurring loop)
2 
3A **scheduled deployment** fires a fresh session on a cron cadence — the agent
4runs without you. Each firing can carry its own outcome, nesting the bounded
5grade→iterate loop inside every recurring run.
6 
7See [`../../references/loops-and-workflows.md`](../../references/loops-and-workflows.md).
8 
9## Choose the trigger
10 
11| Answer | Shape | Tool |
12|---|---|---|
13| "every morning / weekly / nightly" | recurring cron deployment | `deployment_builder.py` + `cron_validator.py` |
14| "when X happens" | event-driven curl (documented, not scheduled) | `deployment_builder.py` (message only) |
15| "only when I ask" | on-demand (no deployment) | none — just re-send a `user.message` |
16 
17## Workflow (recurring)
18 
191. **Validate the schedule.**
20   ```bash
21   python3 scripts/cron_validator.py --cron "0 9 * * *" --timezone Europe/Berlin
22   ```
23   Invalid cron/timezone → exit 1. Read the **DST note**: wall-clock semantics mean
24   spring-forward times are skipped and fall-back times fire twice — avoid
25   02:00–03:00 in DST zones if exactly-once matters.
262. **Build the deployment payload.**
27   ```bash
28   python3 scripts/deployment_builder.py \
29     --sheet ./my-agent/build-sheet.json --agent-id agent_123 --env-id env_456 \
30     --nest-outcome --out ./my-agent/payloads/deployment.json
31   ```
32   `--nest-outcome` includes the rubric so **each firing self-grades**. The tool
33   prints the BYOK curl to create it and to **test it once** with the manual `run`
34…