plain-prompt
Analyze a failed systemd service
Correlate unit state, journal excerpts, dependencies, and restart behavior.
- Revision
- 1
- Verified
- 2026-07-26
Compatibility and paths
GenericWorks with instruction-following chat models.
ChatGPT
Claude Code
Gemini CLI
Trust and provenance
Curated record reviewed 2026-07-26. Results still depend on the supplied context and target environment.
Fill variables
Values stay in this browser tab and are not stored.
Generated asset2 required field(s) must be completed
Objective: Locate the first actionable service failure without recommending blind restarts.
Context: A restart may erase timing evidence or amplify a crash loop.
Instructions:
- Use only the supplied evidence and identify missing information.
- Lead with the finding, confidence, and one safe next action.
- Put read-only checks before mutations and include a stop condition.
- Never request or reproduce secrets, credentials, or private keys.
User request:
Unit and host:
{{environment}}
systemctl/journal evidence:
{{evidence}}
Return cause, confidence, next check, and stop condition.
Required output:
- Finding
- Evidence
- Confidence
- Safe next check
- Stop conditionReal example
Input
nginx.service failed; ExecStartPre exited 1; nginx: [emerg] unexpected '}' in /etc/nginx/nginx.conf:42
Expected result
Cause: configuration parse failure at line 42 before nginx started. Next: inspect the surrounding block and run `nginx -t`; stop before restart until the syntax test passes.