Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Fish shell configuration, aliases, gRPC debugging, PostgreSQL CLI helpers, and dev workflow shortcuts for the Anachak platform.
.claude/skills/iamfourh2e-terminal-fish/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -39% | 0% |
The developer uses fish shell (/opt/homebrew/bin/fish) with a customized set of aliases, functions, and PostgreSQL helpers stored in ~/.config/fish/conf.d/.
| File | Purpose | |---|---| | ~/.config/fish/conf.d/zzz_aliases.fish | Navigation, git, system, gRPC, Rust/Cargo shortcuts | | ~/.config/fish/conf.d/pg.fish | PostgreSQL CLI toolkit with dot notation + JSON output |
| Topic | File | |---|---| | All aliases & shortcuts | aliases.md | | PostgreSQL helpers | postgresql.md | | gRPC debugging | grpc-debugging.md | | Dev workflow patterns | workflow-patterns.md |
fishdg # cd ~/Documents/Github cw # cargo watch -x run --no-restart (hot reload) cc # cargo check ct # cargo test ship "feat: add billing" # git add -A && commit && push origin main kk # kill all dev server ports
fishpguse anachak_dev # switch active database pg auth_users # dump table as JSON pgc payments # row count pgs subscriptions # schema + indexes pgf auth_users admin email # full-text search pgq "SELECT * FROM..." # raw SQL → JSON
fishlist # list all gRPC services show AuthService # list RPCs for a service call '{"identifier":"admin"}' auth.v1.AuthService/Login # call with JSON
$PG_DB universal variable persists the active database across sessions (default: anachak_dev)localhost:50050 (the Anachak Rust backend default port)kk kills ports: 8080-8082, 50050-50054, 5173-5174, 3000-3002, 9090-9092pg* commands pipe JSON output through jq . for pretty-printingpg anachak_dev.auth_users or pgc rust_grpc.payments| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 5,213 | 2,118 | -59% | 1 | 1 | 0% | 1,101 | 994 | -10% | 0 | 0 | — |
case-02 | fail→pass | 6,303 | 3,029 | -52% | 1 | 1 | 0% | 1,239 | 1,184 | -4% | 0 | 0 | — |
case-03 | fail→pass | 4,972 | 3,903 | -22% | 1 | 1 | 0% | 897 | 989 | +10% | 0 | 0 | — |
case-04 | fail→pass | 6,092 | 1,500 | -75% | 1 | 1 | 0% | 1,159 | 877 | -24% | 0 | 0 | — |
case-05 | fail→pass | 10,340 | 2,465 | -76% | 1 | 1 | 0% | 1,640 | 1,000 | -39% | 0 | 0 | — |
case-10 | fail→pass | 7,518 | 2,297 | -69% | 1 | 1 | 0% | 1,429 | 1,029 | -28% | 0 | 0 | — |
case-06 | pass→pass | 5,535 | 1,626 | -71% | 1 | 1 | 0% | 1,002 | 899 | -10% | 0 | 0 | — |
case-07 | pass→pass | 3,548 | 1,418 | -60% | 1 | 1 | 0% | 627 | 837 | +33% | 0 | 0 | — |
case-08 | fail→pass | 4,685 | 1,704 | -64% | 1 | 1 | 0% | 901 | 937 | +4% | 0 | 0 | — |
case-09 | fail→pass | 9,957 | 2,854 | -71% | 1 | 1 | 0% | 1,996 | 1,132 | -43% | 0 | 0 | — |
case-11 | fail→pass | 7,788 | 2,090 | -73% | 1 | 1 | 0% | 1,510 | 1,027 | -32% | 0 | 0 | — |
case-12 | fail→pass | 5,092 | 1,549 | -70% | 1 | 1 | 0% | 958 | 917 | -4% | 0 | 0 | — |
case-13 | fail→pass | 2,770 | 1,397 | -50% | 1 | 1 | 0% | 511 | 836 | +64% | 0 | 0 | — |
case-14 | fail→pass | 8,076 | 1,765 | -78% | 1 | 1 | 0% | 1,618 | 907 | -44% | 0 | 0 | — |
case-15 | fail→pass | 4,572 | 1,441 | -68% | 1 | 1 | 0% | 866 | 861 | -1% | 0 | 0 | — |
case-16 | fail→pass | 5,382 | 1,124 | -79% | 1 | 1 | 0% | 1,026 | 789 | -23% | 0 | 0 | — |
case-17 | fail→pass | 3,749 | 3,399 | -9% | 1 | 1 | 0% | 644 | 874 | +36% | 0 | 0 | — |
case-18 | fail→pass | 4,423 | 1,390 | -69% | 1 | 1 | 0% | 734 | 844 | +15% | 0 | 0 | — |
case-19 | fail→pass | 4,805 | 1,520 | -68% | 1 | 1 | 0% | 894 | 872 | -2% | 0 | 0 | — |
case-20 | fail→pass | 8,288 | 1,481 | -82% | 1 | 1 | 0% | 1,795 | 847 | -53% | 0 | 0 | — |
case-21 | pass→pass | 2,698 | 2,349 | -13% | 1 | 1 | 0% | 539 | 876 | +63% | 0 | 0 | — |
case-22 | pass→pass | 4,682 | 3,422 | -27% | 1 | 1 | 0% | 1,083 | 1,294 | +19% | 0 | 0 | — |
case-23 | pass→pass | 5,251 | 3,977 | -24% | 1 | 1 | 0% | 1,074 | 1,484 | +38% | 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. 23 cases were attempted. The headline lift of +78 percentage points is the difference between those two pass rates over the 23 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.