From junjak-ai-harness
LSP-driven debugging patterns for TypeScript codebases. Complements the general debugging methodology in superpowers:systematic-debugging. Use in Phase 3-4 escalation scenarios or when investigating unexpected behavior.
How this skill is triggered — by the user, by Claude, or both
Slash command
/junjak-ai-harness:debugThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill covers **LSP-accelerated debugging patterns** specific to this harness. For the full systematic debugging methodology (reproduce → narrow → hypothesize → test), use `superpowers:systematic-debugging`.
This skill covers LSP-accelerated debugging patterns specific to this harness. For the full systematic debugging methodology (reproduce → narrow → hypothesize → test), use superpowers:systematic-debugging.
superpowers:systematic-debugging for the reproduction stephover to check inferred types against assumptionsincomingCalls / outgoingCalls to map the flowbunx tsc --noEmit + project test command; confirm no regression| Symptom | First LSP operation | Second (if needed) |
|---|---|---|
X is not a function | goToDefinition on X — is it exported? | findReferences — is everyone importing the same symbol? |
Cannot read property Y of undefined | hover on the object — what's the inferred type? | goToDefinition on the type — does it declare Y? |
| Type mismatch after refactor | findReferences on the changed type | goToImplementation on interfaces — any implementer missed? |
| Downstream code broken unexpectedly | incomingCalls on the changed function | Check each caller manually |
| Stale cache / wrong data | findReferences on store/query key | Confirm every call site uses the same key |
| Unfamiliar function signature | hover for signature | documentSymbol for neighboring helpers |
superpowers:systematic-debugging)bunx tsc --noEmit before declaring the fix completefindReferences to confirm no orphaned callersFor non-TypeScript investigation patterns, fall back to superpowers:systematic-debugging.
Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub junjak/ai-harness --plugin junjak-ai-harness