Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Redirect to claude-embeddings-search for tool use (function calling) and agentic loop patterns with Claude. Use when looking for the secondary Anthropic workflow. Trigger with "anthropic tools", "claude function calling".
.claude/skills/jeremylongshore-clade-core-workflow-b/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -31% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -30% | 0% |
This skill redirects to clade-embeddings-search which covers tool use (function calling), the agentic tool loop, and building Claude-powered agents.
clade-model-inferenceThis skill has been replaced. The secondary Anthropic workflow is tool use / function calling, covered in full by clade-embeddings-search.
clade-embeddings-search| Issue | Solution | |-------|----------| | Skill not found | Run clade-embeddings-search directly | | Tool use errors | See tool validation patterns in that skill |
typescript// Use claude-embeddings-search for the full tool use guide const tools: Anthropic.Tool[] = [{ name: 'get_weather', description: 'Get weather for a city', input_schema: { type: 'object', properties: { city: { type: 'string' } }, required: ['city'], }, }]; const response = await client.messages.create({ model: 'claude-sonnet-4-20250514', max_tokens: 1024, tools, messages: [{ role: 'user', content: "What's the weather in Paris?" }], });
Run clade-embeddings-search for the complete tool use guide.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 15,650 | 15,547 | -1% | 1 | 1 | 0% | 2,957 | 3,586 | +21% | 0 | 0 | — |
case-02 | fail→fail | 15,791 | 14,457 | -8% | 1 | 1 | 0% | 3,103 | 3,183 | +3% | 0 | 0 | — |
case-03 | fail→fail | 17,627 | 16,893 | -4% | 1 | 1 | 0% | 3,611 | 3,731 | +3% | 0 | 0 | — |
case-04 | fail→pass | 9,095 | 5,123 | -44% | 1 | 1 | 0% | 1,431 | 1,446 | +1% | 0 | 0 | — |
case-05 | fail→pass | 9,705 | 4,980 | -49% | 1 | 1 | 0% | 1,833 | 1,364 | -26% | 0 | 0 | — |
case-06 | fail→pass | 12,671 | 8,339 | -34% | 1 | 1 | 0% | 2,449 | 2,013 | -18% | 0 | 0 | — |
case-07 | fail→pass | 12,093 | 6,074 | -50% | 1 | 1 | 0% | 2,283 | 1,566 | -31% | 0 | 0 | — |
case-08 | fail→fail | 19,977 | 9,830 | -51% | 1 | 1 | 0% | 2,746 | 2,322 | -15% | 0 | 0 | — |
case-09 | fail→pass | 11,777 | 5,390 | -54% | 1 | 1 | 0% | 2,137 | 1,502 | -30% | 0 | 0 | — |
case-10 | fail→pass | 14,225 | 5,279 | -63% | 1 | 1 | 0% | 2,482 | 1,319 | -47% | 0 | 0 | — |
case-11 | fail→pass | 11,978 | 3,701 | -69% | 1 | 1 | 0% | 1,853 | 1,075 | -42% | 0 | 0 | — |
case-12 | fail→fail | 12,823 | 8,520 | -34% | 1 | 1 | 0% | 2,861 | 2,248 | -21% | 0 | 0 | — |
case-13 | fail→pass | 11,244 | 7,477 | -34% | 1 | 1 | 0% | 2,048 | 1,748 | -15% | 0 | 0 | — |
case-14 | fail→pass | 10,519 | 4,797 | -54% | 1 | 1 | 0% | 1,909 | 1,360 | -29% | 0 | 0 | — |
case-15 | fail→pass | 8,672 | 2,134 | -75% | 1 | 1 | 0% | 1,379 | 837 | -39% | 0 | 0 | — |
case-16 | fail→pass | 14,713 | 7,309 | -50% | 1 | 1 | 0% | 2,508 | 1,741 | -31% | 0 | 0 | — |
case-17 | fail→pass | 6,819 | 1,846 | -73% | 1 | 1 | 0% | 982 | 778 | -21% | 0 | 0 | — |
case-18 | fail→pass | 10,763 | 2,678 | -75% | 1 | 1 | 0% | 1,989 | 902 | -55% | 0 | 0 | — |
case-19 | fail→pass | 10,543 | 2,646 | -75% | 1 | 1 | 0% | 1,678 | 843 | -50% | 0 | 0 | — |
case-20 | pass→pass | 6,284 | 6,435 | +2% | 1 | 1 | 0% | 1,284 | 1,721 | +34% | 0 | 0 | — |
case-21 | pass→pass | 11,350 | 6,841 | -40% | 1 | 1 | 0% | 2,515 | 2,141 | -15% | 0 | 0 | — |
case-22 | pass→pass | 17,746 | 7,850 | -56% | 1 | 1 | 0% | 2,440 | 2,218 | -9% | 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 +64 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.