Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Master guide for using Claude Code effectively. Includes configuration templates, prompting strategies "Thinking" keywords, debugging techniques, and best practices for interacting with the agent.
.claude/skills/davila7-claude-code-guide/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -42% | 0% |
To provide a comprehensive reference for configuring and using Claude Code (the agentic coding tool) to its full potential. This skill synthesizes best practices, configuration templates, and advanced usage patterns.
CLAUDE.md)When starting a new project, create a CLAUDE.md file in the root directory to guide the agent.
markdown# Project Guidelines ## Commands - Run app: `npm run dev` - Test: `npm test` - Build: `npm run build` ## Code Style - Use TypeScript for all new code. - Functional components with Hooks for React. - Tailwind CSS for styling. - Early returns for error handling. ## Workflow - Read `README.md` first to understand project context. - Before editing, read the file content. - After editing, run tests to verify.
Use these keywords in your prompts to trigger deeper reasoning from the agent:
If the agent is stuck or behaving unexpectedly:
grep or find to locate relevant files first.CLAUDE.md.Based on Claude Code Guide by zebbern.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 22,844 | 8,166 | -64% | 1 | 1 | 0% | 2,181 | 1,514 | -31% | 0 | 0 | — |
case-02 | fail→pass | 14,614 | 10,567 | -28% | 1 | 1 | 0% | 2,354 | 2,203 | -6% | 0 | 0 | — |
case-03 | fail→pass | 16,442 | 16,017 | -3% | 1 | 1 | 0% | 2,456 | 2,816 | +15% | 0 | 0 | — |
case-04 | pass→pass | 12,659 | 2,219 | -82% | 1 | 1 | 0% | 1,596 | 747 | -53% | 0 | 0 | — |
case-05 | pass→pass | 11,465 | 7,241 | -37% | 1 | 1 | 0% | 2,185 | 1,675 | -23% | 0 | 0 | — |
case-06 | pass→pass | 14,250 | 9,491 | -33% | 1 | 1 | 0% | 2,375 | 1,979 | -17% | 0 | 0 | — |
case-07 | pass→pass | 12,414 | 2,070 | -83% | 1 | 1 | 0% | 1,015 | 770 | -24% | 0 | 0 | — |
case-08 | fail→pass | 10,796 | 2,627 | -76% | 1 | 1 | 0% | 1,545 | 880 | -43% | 0 | 0 | — |
case-09 | pass→pass | 11,284 | 3,540 | -69% | 1 | 1 | 0% | 1,716 | 931 | -46% | 0 | 0 | — |
case-10 | pass→pass | 10,363 | 4,188 | -60% | 1 | 1 | 0% | 1,585 | 1,087 | -31% | 0 | 0 | — |
case-11 | pass→pass | 12,323 | 12,220 | -1% | 1 | 1 | 0% | 1,930 | 2,371 | +23% | 0 | 0 | — |
case-12 | pass→pass | 17,983 | 11,686 | -35% | 1 | 1 | 0% | 2,595 | 2,398 | -8% | 0 | 0 | — |
case-13 | fail→pass | 12,658 | 9,200 | -27% | 1 | 1 | 0% | 1,974 | 1,581 | -20% | 0 | 0 | — |
case-14 | pass→pass | 14,783 | 2,527 | -83% | 1 | 1 | 0% | 2,377 | 772 | -68% | 0 | 0 | — |
case-15 | pass→pass | 15,158 | 2,767 | -82% | 1 | 1 | 0% | 2,498 | 880 | -65% | 0 | 0 | — |
case-16 | pass→pass | 16,397 | 14,207 | -13% | 1 | 1 | 0% | 2,472 | 2,572 | +4% | 0 | 0 | — |
case-17 | fail→pass | 10,303 | 3,151 | -69% | 1 | 1 | 0% | 1,543 | 888 | -42% | 0 | 0 | — |
case-18 | fail→pass | 9,587 | 2,150 | -78% | 1 | 1 | 0% | 1,391 | 776 | -44% | 0 | 0 | — |
case-19 | fail→pass | 10,225 | 2,844 | -72% | 1 | 1 | 0% | 1,544 | 840 | -46% | 0 | 0 | — |
case-20 | pass→pass | 8,317 | 5,887 | -29% | 1 | 1 | 0% | 1,590 | 1,595 | +0% | 0 | 0 | — |
case-21 | pass→pass | 14,927 | 13,213 | -11% | 1 | 1 | 0% | 2,367 | 2,561 | +8% | 0 | 0 | — |
case-22 | pass→pass | 10,497 | 9,189 | -12% | 1 | 1 | 0% | 1,852 | 2,193 | +18% | 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 +32 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.