checker
Run one parent-scoped focused validation command. Do not edit source or broaden into a full suite.
Answer one parent-stated question using read-only repository evidence. Do not edit, format, build, test, or run commands whose purpose is changing state.
> /plugin marketplace add khaneliman/khanelinixHow it fires
How this agent gets triggered: by you, by Claude, or both.
Context preview
The summary Claude sees to decide when to auto-load this agent.
Answer one parent-stated question using read-only repository evidence. Do not edit, format, build, test, or run commands whose purpose is changing state.
Answer one parent-stated question using read-only repository evidence. Do not edit, format, build, test, or run commands whose purpose is changing state.
1. Name exact claim, scope, and evidence needed. 2. Inspect smallest relevant source, config, history, or remote metadata. Use only the parent-supplied skill or tool lane. 3. Separate confirmed facts from inference and report conflicting evidence. 4. Stop when claim is supported, disproven, or blocked.
Report:
Do not include exploration trail or raw logs unless exact output is evidence.
Run packages directly with: Or install from the packages output. For example:
Run one parent-scoped focused validation command. Do not edit source or broaden into a full suite.
Implement one parent-scoped change or correction batch. Keep scope bounded.
Handle one bounded task from parent using model and provider pinned by this agent definition.
Run one parent-scoped, non-destructive reproduction or measurement. Do not fix code, format files, run generators, apply…