Llm Integration
This is a comprehensive reference for production LLM integration covering function calling, streaming, local inference with Ollama, and fine-tuning. You get 20 rules across 7 categories with concrete examples: strict mode tool schemas, SSE streaming endpoints, QLoRA training configs, and context optimization patterns. The documentation is opinionated in the right ways, recommending deepseek-r1:70b for local reasoning, 5-15 tools max per request, and trying prompting before fine-tuning. Most useful when you're moving beyond basic API calls and need battle-tested patterns for tool use, real-time responses, or cost optimization through local models. The quick reference tables and code snippets make this genuinely reusable rather than just theoretical guidance.
- skills.sh installs
- 216
- GitHub stars
- 301
- last updated
- Aug 21, 2026
Install
$ npx -y skills add yonatangross/orchestkit --skill llm-integration --agent claude-codeInstalls into .claude/skills of the current project.
Files
LLM Integration
Patterns for integrating LLMs into production applications: tool use, streaming, local inference, and fine-tuning. Each category has individual rule files in rules/ loaded on-demand.
Quick Reference
| Category | Rules | Impact | When to Use |
|---|---|---|---|
| Function Calling | 3 | CRITICAL | Tool definitions, parallel execution, input validation |
| Streaming | 3 | HIGH | SSE endpoints, structured streaming, backpressure handling |
| Local Inference | 3 | HIGH | Ollama setup, model selection, GPU optimization |
| Fine-Tuning | 3 | HIGH | LoRA/QLoRA training, dataset preparation, evaluation |
| Context Optimization | 2 | HIGH | Window management, compression, caching, budget scaling |
| Evaluation | 2 | HIGH | LLM-as-judge, RAGAS metrics, quality gates, benchmarks |
| Prompt Engineering | 4 | HIGH | CoT, few-shot, versioning, DSPy optimization, ReAct, cost optimization |
Total: 20 rules across 7 categories
Quick Start
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageDetails
- Repository
- github.com/yonatangross/orchestkit
- Category
- AI & Agent Building
- First seen
- Jun 3, 2026
More from yonatangross/orchestkit
All 61- I18n Date Patterns215
- Task Dependency Patterns214
- Api Design211
- Assess211
- Analytics210
- Commit210
- Skill Evolution210
- Verify210
- Create Pr209
- Audit Full206
- Implement206
- Configure196
- Demo Producer195
- Async Jobs194
- Golden Dataset193
- Product Frameworks193
- Browser Tools191
- Mcp Patterns190
- Feedback180
- Scope Appropriate Architecture179
- Testing Patterns179
- Help166
- Checkpoint Resume156
- Dashboard Patterns124
Recommended
More AI & Agent Building
self-improving-agentkimasplund/clawdbot-skills-packEnables continuous self-improvement through learning from failures, user corrections, and capability gaps. Integrates with QAVR for learned memory ranking.212skills.sh installs
prompt-optimizerchujianyun/skillsPrompt 优化助手。适用于用户想优化提示词、改进 AI 指令、为特定任务设计更好的 prompt，或需要选择合适提示框架时使用。会根据任务场景匹配合适框架，必要时先追问关键信息，再输出更清晰、更可执行的提示词版本。211skills.sh installs
ai-elementsexistential-birds/beagleVercel AI Elements for workflow UI components. Use when building chat interfaces, displaying tool execution, showing reasoning/thinking, or creating job queues.209skills.sh installs
user-interface-designerqodex-ai/ai-agent-skillsDesign user-centric interfaces for web applications. Creates wireframes, visual designs, and interactive prototypes with usability focus.206skills.sh installs