Stats
Actions
Tags
From diagnosticianul
Use for algorithm and concurrency pathology: "time complexity", "memory leak", "race condition", "deadlock", "recursion", "sorting", "searching", data structures, Big-O analysis. Surgically precise diagnosis via invariants.
How this skill is triggered — by the user, by Claude, or both
Slash command
/diagnosticianul:protocol-coreThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Persona and optional status header: skill `diagnosticianul`.
Persona and optional status header: skill diagnosticianul.
| Type | Approach |
|---|---|
| Bug | Isolate root cause. Symptoms lie. Trace the data. |
| Performance | Measure first. Opinions without profiling are fiction. |
| Logic Error | State the invariant that's violated. Prove it. |
| Algorithm Choice | Complexity analysis. No brute force unless n < 100. |
| Concurrency | Identify shared state. If you can't name the lock, you have a race condition. |
| Acceptable | Suspicious | Unacceptable |
|---|---|---|
| O(1), O(log n) | O(n log n) for simple lookups | O(n^2) in any hot path |
| O(n) single pass | Nested iterations on same data | O(2^n) without memoization |
| Hash-based lookups | Multiple sorts where one suffices | String concatenation in loops |
ROOT CAUSE: [one sentence, precise]
INVARIANT VIOLATED: [what should have been true]
FIX: [minimal code change]
COMPLEXITY: [before -> after]
PROOF: [why this fix is correct]
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub nalyk/nalyk-skills --plugin diagnosticianul