docs(scouts): scrub host-TZ/DST reference from README

[theagenticguy]

Summary

The README's cron section leaked internal ops context ("if your devbox runs UTC and you want local-CST time, convert accordingly… DST will shift the run by an hour twice a year") — that's our own setup talking to itself, not scheduler-agnostic user docs.

Change

Replace with a scheduler-neutral section that lists cron, systemd timers, GitHub Actions schedule:, and launchd as equally valid options, and keeps only the genuinely useful guidance (pin RALPH_USER_DIR in the entry; seed a scout before the first run).

Test plan

Docs-only; no code paths touched.

[github-actions]

Dependency Review

✅ No vulnerabilities or license issues or OpenSSF Scorecard issues found.

Scanned Files

None

[Copilot]

Pull request overview

Updates the scouts/ README to remove internal host timezone/DST guidance and replace it with a scheduler-neutral “run on a schedule” section that applies across environments.

Changes:

• Renames the cron-specific section to a scheduler-agnostic “Running on a schedule”.
• Broadens scheduling options mentioned (cron/systemd/GitHub Actions/launchd).
• Updates the example and guidance around setting RALPH_USER_DIR and seeding a scout before scheduled runs.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The list of “any scheduler works — … GitHub Actions schedule: …” is a bit misleading given this README’s earlier guidance that scouts/ is user data and not committed. A GitHub Actions runner is typically ephemeral and won’t retain scouts/ between runs, so scheduled runs won’t work unless you also document how to provision/seed scouts on every run (or where they come from). Consider either removing GitHub Actions from the “equally valid” list or adding a short qualifier about needing to create/provide scouts on the runner each run.

[Copilot]

The text says “Pin RALPH_USER_DIR in the entry”, but the example sets RALPH_USER_DIR on its own line. That standalone assignment only applies if it’s placed as an environment setting line in the crontab (not inside a script), so readers may paste this into the wrong place. Consider either putting RALPH_USER_DIR=... inline on the 0 9 * * * ... line, or explicitly label it as a crontab-wide env var line.

Suggested change

	RALPH_USER_DIR=$HOME/ralph-data
	0 9 * * * /path/to/ralph-for-kiro scout \
	0 9 * * * RALPH_USER_DIR=$HOME/ralph-data /path/to/ralph-for-kiro scout \

[Copilot]

“The scheduler will fail silently if scouts/ is empty” isn’t quite accurate: ralph scout logs an error (“No scouts found…”) and returns, and it’s the scheduled run (not the scheduler) that does nothing. Consider rewording to something like “Scheduled runs will exit without doing work if no scouts exist; run ralph scout init … first”, and (if relevant) note that redirecting output is how you avoid missing the error in cron.

Suggested change

	The scheduler will fail silently if `scouts/` is empty — seed with
	`ralph scout init` or `--from-example` once before the first run.
	Scheduled runs will exit without doing work if no scouts exist — run
	`ralph scout init <name>` or `ralph scout init --from-example <template> <name>`
	once before the first scheduled run.