Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Modern file and content search using fd, ripgrep (rg), and fzf. Triggers on: fd, ripgrep, rg, find files, search code, fzf, fuzzy find, search codebase.
.claude/skills/aiskillstore-file-search/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 39% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 93% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 0% | 0% |
Modern file and content search.
bash# Find by name fd config # Files containing "config" fd -e py # Python files # By type fd -t f config # Files only fd -t d src # Directories only # Exclude fd -E node_modules # Exclude directory fd -E "*.min.js" # Exclude pattern # Execute command fd -e py -x wc -l # Line count per file
bash# Simple search rg "TODO" # Find TODO rg -i "error" # Case-insensitive # By file type rg -t py "import" # Python files only rg -t js -t ts "async" # JS and TS # Context rg -C 3 "function" # 3 lines before/after # Output modes rg -l "TODO" # File names only rg -c "TODO" # Count per file
bash# Find and select fd | fzf # With preview fd | fzf --preview 'bat --color=always {}' # Multi-select fd -e ts | fzf -m | xargs code
bash# Find files, search content fd -e py -x rg "async def" {} # Search, select, open rg -l "pattern" | fzf --preview 'rg -C 3 "pattern" {}' | xargs vim
| Task | Command | |------|---------| | Find TS files | fd -e ts | | Find in src | fd -e ts src/ | | Search pattern | rg "pattern" | | Search in type | rg -t py "import" | | Files with match | rg -l "pattern" | | Count matches | rg -c "pattern" | | Interactive | fd \| fzf | | With preview | fd \| fzf --preview 'bat {}' |
| Tip | Why | |-----|-----| | Both respect .gitignore | Auto-skip node_modules, dist | | Use -t over -g | Type flags are faster | | Narrow the path | rg pattern src/ faster | | Use -F for literals | Avoids regex overhead |
For detailed patterns, load:
./references/advanced-workflows.md - Git integration, shell functions, power workflows| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 3,420 | 2,191 | -36% | 1 | 1 | 0% | 440 | 848 | +93% | 0 | 0 | — |
case-02 | pass→pass | 7,315 | 4,121 | -44% | 1 | 1 | 0% | 1,156 | 1,151 | -0% | 0 | 0 | — |
case-03 | pass→pass | 5,487 | 2,741 | -50% | 1 | 1 | 0% | 868 | 1,015 | +17% | 0 | 0 | — |
case-04 | pass→pass | 6,299 | 2,111 | -66% | 1 | 1 | 0% | 1,034 | 923 | -11% | 0 | 0 | — |
case-05 | pass→pass | 4,158 | 2,079 | -50% | 1 | 1 | 0% | 627 | 870 | +39% | 0 | 0 | — |
case-06 | pass→pass | 6,846 | 3,256 | -52% | 1 | 1 | 0% | 1,080 | 1,025 | -5% | 0 | 0 | — |
case-07 | pass→pass | 4,462 | 2,357 | -47% | 1 | 1 | 0% | 704 | 912 | +30% | 0 | 0 | — |
case-08 | pass→pass | 4,693 | 1,994 | -58% | 1 | 1 | 0% | 708 | 868 | +23% | 0 | 0 | — |
case-09 | pass→pass | 2,165 | 1,626 | -25% | 1 | 1 | 0% | 312 | 812 | +160% | 0 | 0 | — |
case-10 | fail→pass | 6,277 | 2,735 | -56% | 1 | 1 | 0% | 1,104 | 990 | -10% | 0 | 0 | — |
case-11 | pass→pass | 7,428 | 3,535 | -52% | 1 | 1 | 0% | 1,340 | 1,072 | -20% | 0 | 0 | — |
case-12 | pass→pass | 9,832 | 5,821 | -41% | 1 | 1 | 0% | 1,666 | 1,540 | -8% | 0 | 0 | — |
case-13 | pass→pass | 8,947 | 3,410 | -62% | 1 | 1 | 0% | 1,556 | 1,171 | -25% | 0 | 0 | — |
case-14 | fail→pass | 5,769 | 3,838 | -33% | 1 | 1 | 0% | 896 | 1,249 | +39% | 0 | 0 | — |
case-15 | fail→pass | 12,882 | 4,273 | -67% | 1 | 1 | 0% | 2,241 | 1,250 | -44% | 0 | 0 | — |
case-16 | pass→pass | 8,348 | 3,537 | -58% | 1 | 1 | 0% | 1,236 | 1,047 | -15% | 0 | 0 | — |
case-17 | pass→pass | 14,989 | 7,697 | -49% | 1 | 1 | 0% | 2,340 | 1,663 | -29% | 0 | 0 | — |
case-18 | pass→pass | 14,787 | 7,168 | -52% | 1 | 1 | 0% | 1,930 | 1,702 | -12% | 0 | 0 | — |
case-19 | pass→pass | 2,737 | 2,715 | -1% | 1 | 1 | 0% | 389 | 957 | +146% | 0 | 0 | — |
case-20 | pass→pass | 6,021 | 4,613 | -23% | 1 | 1 | 0% | 1,046 | 1,285 | +23% | 0 | 0 | — |
case-21 | pass→pass | 7,226 | 5,668 | -22% | 1 | 1 | 0% | 1,284 | 1,602 | +25% | 0 | 0 | — |
case-22 | pass→pass | 7,721 | 4,377 | -43% | 1 | 1 | 0% | 1,330 | 1,322 | -1% | 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. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.