Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when the user is debugging a bug, tracing an error, or asking why something fails. Examples: "Why is X failing?", "Where does this error come from?", "Trace this bug"
.claude/skills/reinamaccredy-gitnexus-debugging/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -37% | 0% |
1. gitnexus_query({query: "<error or symptom>"}) → Find related execution flows
2. gitnexus_context({name: "<suspect>"}) → See callers/callees/processes
3. READ gitnexus://repo/{name}/process/{name} → Trace execution flow
4. gitnexus_cypher({query: "MATCH path..."}) → Custom traces if needed> If "Index is stale" → run npx gitnexus analyze in terminal.
- [ ] Understand the symptom (error message, unexpected behavior)
- [ ] gitnexus_query for error text or related code
- [ ] Identify the suspect function from returned processes
- [ ] gitnexus_context to see callers and callees
- [ ] Trace execution flow via process resource if applicable
- [ ] gitnexus_cypher for custom call chain traces if needed
- [ ] Read source files to confirm root cause| Symptom | GitNexus Approach | | -------------------- | ---------------------------------------------------------- | | Error message | gitnexus_query for error text → context on throw sites | | Wrong return value | context on the function → trace callees for data flow | | Intermittent failure | context → look for external calls, async deps | | Performance issue | context → find symbols with many callers (hot paths) | | Recent regression | detect_changes to see what your changes affect |
gitnexus_query — find code related to error:
gitnexus_query({query: "payment validation error"})
→ Processes: CheckoutFlow, ErrorHandling
→ Symbols: validatePayment, handlePaymentError, PaymentExceptiongitnexus_context — full context for a suspect:
gitnexus_context({name: "validatePayment"})
→ Incoming calls: processCheckout, webhookHandler
→ Outgoing calls: verifyCard, fetchRates (external API!)
→ Processes: CheckoutFlow (step 3/7)gitnexus_cypher — custom call chain traces:
cypherMATCH path = (a)-[:CodeRelation {type: 'CALLS'}*1..2]->(b:Function {name: "validatePayment"}) RETURN [n IN nodes(path) | n.name] AS chain
1. gitnexus_query({query: "payment error handling"})
→ Processes: CheckoutFlow, ErrorHandling
→ Symbols: validatePayment, handlePaymentError
2. gitnexus_context({name: "validatePayment"})
→ Outgoing calls: verifyCard, fetchRates (external API!)
3. READ gitnexus://repo/my-app/process/CheckoutFlow
→ Step 3: validatePayment → calls fetchRates (external)
4. Root cause: fetchRates calls external API without proper timeout| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | fail→fail | 10,982 | 5,614 | -49% | 1 | 1 | 0% | 1,548 | 1,078 | -30% | 0 | 0 | — |
case-09 | fail→fail | 15,809 | 8,395 | -47% | 1 | 1 | 0% | 2,182 | 1,332 | -39% | 0 | 0 | — |
case-01 | fail→fail | 6,662 | 4,874 | -27% | 1 | 1 | 0% | 257 | 984 | +283% | 0 | 0 | — |
case-02 | fail→fail | 7,052 | 16,219 | +130% | 1 | 1 | 0% | 198 | 940 | +375% | 0 | 0 | — |
case-10 | fail→pass | 13,856 | 2,610 | -81% | 1 | 1 | 0% | 1,209 | 1,067 | -12% | 0 | 0 | — |
case-03 | fail→fail | 19,728 | 5,788 | -71% | 1 | 1 | 0% | 2,657 | 979 | -63% | 0 | 0 | — |
case-04 | fail→pass | 11,566 | 2,577 | -78% | 1 | 1 | 0% | 1,935 | 1,083 | -44% | 0 | 0 | — |
case-05 | fail→pass | 11,359 | 4,600 | -60% | 1 | 1 | 0% | 1,972 | 1,504 | -24% | 0 | 0 | — |
case-06 | fail→fail | 16,585 | 6,583 | -60% | 1 | 1 | 0% | 2,542 | 1,041 | -59% | 0 | 0 | — |
case-07 | fail→fail | 19,926 | 8,423 | -58% | 1 | 1 | 0% | 2,607 | 1,081 | -59% | 0 | 0 | — |
case-11 | fail→pass | 9,113 | 2,371 | -74% | 1 | 1 | 0% | 1,534 | 1,104 | -28% | 0 | 0 | — |
case-12 | pass→pass | 18,787 | 7,387 | -61% | 1 | 1 | 0% | 1,570 | 1,727 | +10% | 0 | 0 | — |
case-13 | fail→pass | 22,429 | 8,368 | -63% | 1 | 1 | 0% | 3,416 | 2,139 | -37% | 0 | 0 | — |
case-14 | fail→pass | 15,435 | 67,398 | +337% | 1 | 1 | 0% | 1,801 | 1,613 | -10% | 0 | 0 | — |
case-15 | fail→pass | 44,723 | 1,679 | -96% | 1 | 1 | 0% | 4,154 | 992 | -76% | 0 | 0 | — |
case-16 | fail→fail | 20,658 | 5,511 | -73% | 1 | 1 | 0% | 2,844 | 999 | -65% | 0 | 0 | — |
case-17 | fail→pass | 10,490 | 3,364 | -68% | 1 | 1 | 0% | 1,451 | 1,325 | -9% | 0 | 0 | — |
case-18 | fail→pass | 14,744 | 13,946 | -5% | 1 | 1 | 0% | 2,491 | 2,160 | -13% | 0 | 0 | — |
case-19 | pass→pass | 10,838 | 8,351 | -23% | 1 | 1 | 0% | 1,453 | 1,897 | +31% | 0 | 0 | — |
case-20 | pass→fail | 14,422 | 4,370 | -70% | 1 | 1 | 0% | 2,293 | 888 | -61% | 0 | 0 | — |
case-21 | fail→fail | 17,878 | 15,718 | -12% | 1 | 1 | 0% | 2,466 | 2,882 | +17% | 0 | 0 | — |
case-22 | pass→pass | 4,863 | 4,302 | -12% | 1 | 1 | 0% | 705 | 1,292 | +83% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 14 counted toward the lift figure. The other 8 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +36 percentage points is the difference between those two pass rates over the 14 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.