One Integration at a Time
Every integration (email, calendar, messaging, web scraping) is a separate failure mode with its own configuration, credentials, and edge cases. Trying to set up everything at once leads to debugging five things simultaneously with no working baseline.
The Pattern That Works
- Pick one workflow (e.g., a morning briefing that checks email)
- Get it working end-to-end — config, credentials, test run, verify output
- Add a cron so it runs automatically
- Add error handling (what happens when the email server is down?)
- Move to the next integration
Why This Order Matters
Each step builds on the previous one. You can't debug email integration failures if you're also debugging calendar integration failures at the same time. A working baseline gives you confidence that new failures are caused by the new thing, not something that was already broken.
When Things Break
Run openclaw doctor --fix — it catches most config issues.
For persistent failures, check:
- Credentials (expired tokens, rotated passwords)
- Network (API endpoint changed, firewall rules)
- Rate limits (especially on free-tier APIs)
- Config drift (settings that worked last week but were overwritten by an update)