OpenClaw
Effective troubleshooting starts by turning a vague problem into a reproducible observation. Record what happened, what was expected, when it began, and what changed immediately beforehand.
1. Establish the boundary
Confirm the affected workflow, environment, user, and integration. A narrow boundary makes it easier to separate configuration problems from runtime, connectivity, or tool issues.
2. Inspect evidence in context
Review relevant status output and logs without exposing credentials or unrelated private data. Compare current configuration with the last known working state.
3. Change one variable
Apply the smallest safe correction, then repeat the original workflow. Avoid bundling unrelated changes because doing so makes the result difficult to explain or reverse.
4. Verify the real outcome
A successful command is useful evidence, but the final test is the user-facing workflow. Document the cause, repair, verification result, and any remaining follow-up.
