Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Diagnose and fix TypeScript/language type errors using LSP diagnostics, hover, and code actions. For each type error, fetches the expected vs actual types, surfaces quickfix suggestions, and optionally applies them. Use when getDiagnostics shows type errors you want to resolve quickly.
.claude/skills/bilal140202-ide-type-mismatch-fix/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 65% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-21 | ✗→✓ | ▲ Improved | -33% | 0% |
getToolCapabilities is available.extensionConnected is false: show the same message and stop. If true: proceed.Systematically diagnose and fix type errors using LSP intelligence, without guessing or making assumptions about types.
$ARGUMENTSgetDiagnostics with that urigetDiagnostics with no uri (workspace-wide)error only. If zero errors: report "No type errors found" and stop.For each error (up to 20 total across all files):
getHover at the error position (file, line, column) to fetch:getCodeActions at the error range to get quickfix suggestions from the language servergetSignatureHelp at the call site to see the expected parameter types Error: [error message] File: [file]:[line]:[column] Symbol: [symbol from hover] Actual type: [type from hover] Expected type: [inferred from error message + context] Quickfixes available: [list from getCodeActions] Recommended fix: [your recommendation]
any, fixes an import):previewCodeAction to show the exact diffapplyCodeAction to apply itgetDiagnostics again to verify error count dropped## Type Error Fix Report
Files with errors: N
Total errors processed: M
Fixes applied: K
Remaining errors: R
### Fixed
- src/foo.ts:42 — `string` not assignable to `number` → applied: add type cast
- ...
### Remaining (manual fix needed)
| File | Line | Error | Why manual |
|------|------|-------|-----------|
| src/bar.ts | 17 | ... | Complex generic constraint — see diagnosis below |
| ...
### Diagnoses for remaining errors
[Detailed explanation of each unfixed error with recommended fix]getCodeActions returns no quickfixes and the error involves generics: explain the constraint in plain English from the hover output and stop| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→fail | 3,663 | 2,124 | -42% | 1 | 1 | 0% | 286 | 948 | +231% | 0 | 0 | — |
case-04 | pass→fail | 10,346 | 3,141 | -70% | 1 | 1 | 0% | 2,591 | 1,319 | -49% | 0 | 0 | — |
case-05 | pass→fail | 6,395 | 2,904 | -55% | 1 | 1 | 0% | 1,312 | 1,328 | +1% | 0 | 0 | — |
case-01 | fail→fail | 8,126 | 2,154 | -73% | 1 | 1 | 0% | 1,351 | 933 | -31% | 0 | 0 | — |
case-02 | fail→fail | 5,529 | 2,121 | -62% | 1 | 1 | 0% | 447 | 898 | +101% | 0 | 0 | — |
case-06 | pass→fail | 1,913 | 4,094 | +114% | 1 | 1 | 0% | 208 | 1,579 | +659% | 0 | 0 | — |
case-07 | pass→pass | 8,057 | 1,803 | -78% | 1 | 1 | 0% | 1,420 | 1,130 | -20% | 0 | 0 | — |
case-08 | fail→fail | 9,162 | 2,736 | -70% | 1 | 1 | 0% | 1,817 | 1,281 | -29% | 0 | 0 | — |
case-09 | fail→pass | 7,256 | 1,539 | -79% | 1 | 1 | 0% | 1,407 | 1,097 | -22% | 0 | 0 | — |
case-10 | pass→pass | 7,528 | 1,763 | -77% | 1 | 1 | 0% | 1,304 | 1,076 | -17% | 0 | 0 | — |
case-11 | fail→fail | 4,402 | 1,564 | -64% | 1 | 1 | 0% | 626 | 983 | +57% | 0 | 0 | — |
case-12 | fail→pass | 4,513 | 2,561 | -43% | 1 | 1 | 0% | 755 | 1,245 | +65% | 0 | 0 | — |
case-13 | fail→fail | 10,887 | 1,415 | -87% | 1 | 1 | 0% | 2,107 | 1,034 | -51% | 0 | 0 | — |
case-14 | fail→fail | 2,508 | 3,515 | +40% | 1 | 1 | 0% | 490 | 1,404 | +187% | 0 | 0 | — |
case-15 | fail→fail | 5,108 | 1,855 | -64% | 1 | 1 | 0% | 920 | 1,102 | +20% | 0 | 0 | — |
case-16 | pass→pass | 8,556 | 2,744 | -68% | 1 | 1 | 0% | 1,522 | 1,239 | -19% | 0 | 0 | — |
case-17 | fail→pass | 11,776 | 2,999 | -75% | 1 | 1 | 0% | 2,155 | 1,331 | -38% | 0 | 0 | — |
case-18 | pass→pass | 6,554 | 1,237 | -81% | 1 | 1 | 0% | 1,271 | 1,034 | -19% | 0 | 0 | — |
case-19 | fail→pass | 9,900 | 1,051 | -89% | 1 | 1 | 0% | 1,531 | 957 | -37% | 0 | 0 | — |
case-20 | fail→fail | 4,327 | 1,231 | -72% | 1 | 1 | 0% | 729 | 1,009 | +38% | 0 | 0 | — |
case-21 | fail→pass | 9,942 | 2,441 | -75% | 1 | 1 | 0% | 1,735 | 1,166 | -33% | 0 | 0 | — |
case-22 | fail→pass | 7,802 | 1,302 | -83% | 1 | 1 | 0% | 1,496 | 1,050 | -30% | 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 19 counted toward the lift figure. The other 3 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 +14 percentage points is the difference between those two pass rates over the 19 comparable cases. 3 cases got worse with the skill loaded, and they are 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.