Diagnoses common Linux service issues using logs, systemd/PM2, file permissions, Nginx reverse proxy checks, and DNS sanity checks.
SkillScore100Maintained
Automatic
◆DevOps & SRE pack
Last verified 8/19/2026
What it changes
Passed 23 of 23 test cases with the skill, and 18 of 23 without it.
Without skill78%
With skill100%
Checks it passed that the model alone missed
The response formats the diagnostic output inside a TRIAGE REPORT block structure.
The response explains that kernel tracing and eBPF profiling fall outside standard service triage.
The response asks the user to provide status output or recent logs before recommending fix actions.
and 2 more in the full report
5 fixed18 passed either way0 still failing0 broke
Pass-rate lift
+22 pp
Tokens
+12%
Agent turns
0%
Response time
-26%
Model gemini-3.6-flashMeasured 8/19/2026v123 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 5
Prompt
I need to analyze eBPF stack traces and kernel scheduling latency on Linux kernel 6.1 to debug high CPU wait times. Can you run perf record and profile the scheduler?
What was checked, without and with the skill
Checked
Without skill
With skill
The response states that deep kernel debugging and performance profiling are outside the scope of basic service triage.
Missed
Passed
case-04 · gemini-3.6-flash · 8/19/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md75 lines2 files
Linux & service basics: logs, systemd/PM2, permissions, Nginx reverse proxy, DNS checks
PURPOSE
Diagnoses common Linux service issues using logs, systemd/PM2, file permissions, Nginx reverse proxy checks, and DNS sanity checks.
WHEN TO USE
TRIGGERS:
Show me why this service is failing using logs, then give the exact fix commands.
Restart this app cleanly and confirm it is listening on the right port.
Fix the permissions on this folder so the service can read and write safely.
Set up Nginx reverse proxy for this port and verify DNS and TLS are sane.
Create a systemd service for this script and make it survive reboots.
DO NOT USE WHEN…
You need kernel debugging or deep performance profiling.
You want to exploit systems or bypass access controls.
INPUTS
REQUIRED:
Service type: systemd unit name or PM2 process name.
Observed symptom: error message, status output, or logs (pasted by user).
If explicitly requested and safe: exact shell commands to apply the fix.
Success = service runs, listens on expected port, and reverse proxy/DNS path is correct.
WORKFLOW
Confirm scope and safety:
identify service name and whether changes are permitted.
Gather evidence:
status output + recent logs (see references/triage-commands.md).
Classify failure:
config error, dependency missing, permission denied, port conflict, upstream unreachable, DNS mismatch.