APIs and forms often drift when validation rules are copied between layers. This skill gives Claude, Codex, and Claude Code practical patterns for shared schemas, clear errors, and safer input handling.

Database work often slows down when queries, indexes, and migrations are hard to reason about. This skill gives Claude, Codex, and Claude Code structured SQL guidance for safer and faster database changes.

Custom browser charts often require many D3 decisions before any data appears. This skill supplies chart guidance, templates, helpers, and examples for Claude, Codex, and Claude Code.

Browser restrictions and cross-domain funnels cause advertising platforms to miss purchases. This skill designs a durable capture, reporting, deduplication, and verification workflow.

Production RAG systems need reliable ingestion, updates, filtering, and context recovery. This skill provides patterns for semantic chunking, Qdrant indexing, and filtered retrieval.

Obsidian Bases can hold useful project and note metadata, but querying them manually is slow. This skill gives Claude, Codex, and Claude Code focused prompts for listing bases, querying views, and reading schemas through the local plugin RPC.

Large object arrays are expensive to process through repeated natural-language instructions. This skill applies reusable JavaScript transformations and returns compact processing statistics.

Financial valuation requires consistent assumptions, cash flow projections, and risk testing. This skill provides reusable DCF and sensitivity tools for structured investment analysis.

Financial statements can hide important performance, liquidity, leverage, and valuation signals. This skill calculates key ratios and explains results using benchmarks and trends.

Marketing teams often struggle to compare campaign performance across platforms. This skill calculates engagement, ROI, benchmarks, and recommendations from provided campaign data.

Data science projects often fail when experiments, features, and models lack clear statistical structure. This skill gives Claude, Codex, and Claude Code guided workflows for experimentation, feature engineering, and model review.

Production data teams need reliable patterns for pipelines, quality checks, and orchestration. This skill guides architecture, scaffolding, and DataOps practices for modern data stacks.

Computer vision teams need reliable guidance for models, datasets, inference, and deployment. This skill helps plan and optimize production vision pipelines with practical engineering checks.

Duplicate and near-duplicate images can distort dataset quality and model training. This skill guides Claude, Codex, or Claude Code through FiftyOne Brain workflows to find, review, and remove redundant samples.

Image datasets can hide clusters, outliers, and label problems. This skill guides Claude, Codex, and Claude Code through FiftyOne embedding visualizations.

Computer vision teams need repeatable ways to load media, labels, and predictions into FiftyOne. This skill guides Claude, Codex, and Claude Code through dataset creation, import validation, and model inference.

Dataset imports often fail when media, annotations, and multimodal folders use different formats. This skill guides scanning, detection, package setup, import, grouping, and validation in FiftyOne.

Database changes can cause data loss, slow queries, or access-control errors. This skill provides structured guidance for CookMode V2 schemas, migrations, RLS, and Supabase troubleshooting.

Database work slows down when table names and relationships are unclear. This skill maps Empathy Ledger tables and provides Supabase query patterns for faster planning.

SurrealDB projects often need one guide for documents, graph traversal, vectors, and live queries. This skill gives Python-focused patterns for designing, querying, and tuning multi-model database features.