Optimize Gemini API costs WITHOUT changing the extraction model. Use when building or modifying AI extraction pipelines, batch processing, or when API costs are increasing. Covers caching, prompt optimization, batching, retry logic, and cost tracking. The current model (gemini-3-flash-preview) is proven for PDF invoice extraction and should NOT be downgraded.
This listing is imported from SkillsMP metadata and should be treated as untrusted until upstream source review is completed.
Install skill "cost-aware-llm-pipeline" with this command: npx skills add hmartinezc/skillsmp-hmartinezc-hmartinezc-cost-aware-llm-pipeline
This source entry does not include full markdown content beyond metadata.
This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.
Related by shared tags or category signals.
Safe, idempotent database migration patterns for libSQL/SQLite. Use when modifying server/schema.ts, adding tables or columns, writing seed data, or planning schema changes. Covers idempotent DDL, rollback strategies, and zero-downtime patterns.
Deployment strategies for Node.js fullstack applications using Coolify, Docker, and self-hosted infrastructure. Use when deploying, configuring production environments, troubleshooting deployment issues, or setting up CI/CD pipelines.
Use when: replicar la arquitectura del Smart Logistics Extractor sin IA en cualquier nuevo proyecto transaccional, crear un sistema web con React + Hono + libSQL/Turso, validar que otro repo sigue el mismo patron arquitectonico, preparar despliegue en Coolify o Hetzner.
Validate that changes to AI prompts, extraction logic, or data schemas don't break existing extraction quality. Use when modifying agentPrompts.ts, extractionSchema.ts, or ai.ts routes. Covers golden test sets, diff testing, and confidence score regression.