Teams need reliable WeChat public data for content and account analysis. This skill queries MaxHub endpoints for Channels and Official Accounts and summarizes results.

TikTok research often needs many video, user, search, and ad endpoints. This skill routes requests through MaxHub references and returns structured summaries.

Teams need structured social media data without hand-building API calls. This skill guides Claude, Codex, and Claude Code through MaxHub API lookups across supported platforms.

Teams need a consistent way to parse public video links across supported platforms. This skill guides MaxHub HybridParser calls with clear limits and API key handling.

Startup teams need financial plans that connect revenue, hiring, expenses, runway, and fundraising assumptions. This skill guides Claude, Codex, and Claude Code through structured 3-5 year projections and scenario planning.

Slow SQL queries increase latency and database costs. This skill guides query plans, indexes, pagination, batching, and maintenance choices.

Slow Spark jobs waste cluster budget and delay data delivery. This skill guides Claude, Codex, and Claude Code through partitioning, shuffle, memory, join, caching, and monitoring improvements.

Vector search projects need clear choices for indexes, databases, filters, and ranking. This skill provides practical templates for semantic retrieval systems.

PostgreSQL schemas can fail when data types, constraints, and indexes are chosen without workload context. This skill guides table design with PostgreSQL-specific rules for durable, performant, and maintainable schemas.

Market sizing is hard when sources, assumptions, and methods are scattered. This skill guides Claude, Codex, and Claude Code through defensible TAM, SAM, and SOM analysis.

LLM teams need evidence before changing prompts, models, or retrieval flows. This skill guides automated metrics, judge reviews, human labels, A/B tests, and regression checks.

Embedding choices affect retrieval quality, latency, and cost. This skill helps teams select models, chunk content, and evaluate search results for RAG systems.

Raw metrics often fail to guide stakeholder decisions. This skill turns analysis into clear narratives, visual structures, and action plans.

Unreliable tables can break analytics, reporting, and downstream products. This skill helps define Great Expectations suites, dbt tests, data contracts, and automated validation pipelines.

Trading strategy tests often overstate performance because they ignore bias, costs, and validation discipline. This skill guides Claude, Codex, and Claude Code through robust backtesting patterns.

Financial agents need reliable market, fund, bond, document, news, and macroeconomic data. This skill routes requests to Wind MCP tools and returns sourced results.

Image review tasks need clear scene descriptions, object lists, and extracted text. This skill shows how to call the verging.ai GPT-4o Vision API with image URLs or file uploads.

Large JSON payloads waste tokens and hide important facts. This skill converts JSON or jsonb exports into compact summaries for Claude, Codex, and Claude Code.

Browser apps need a clear CloudBase relational database setup path. This skill shows the shared web SDK client and Supabase-style query flow.

CloudBase MySQL operations need clear tool choice and safety checks. This skill guides agents through MCP-based provisioning, querying, schema changes, and permission reviews.