Large vector databases can become slow, memory-heavy, and hard to benchmark. This skill provides a structured AgentDB optimization workflow for quantization, HNSW indexing, caching, and validation.

Reinforcement learning projects need structured setup, training loops, evaluation, and deployment checks. This skill guides AgentDB users through algorithm selection, training, benchmarking, and production rollout.

Teams need relevant document retrieval for RAG and knowledge bases. This skill guides AgentDB setup, embedding storage, indexing, API search, and relevance tuning.

AI agents often lose useful context between sessions. This skill gives a structured AgentDB workflow for session memory, long-term storage, and pattern learning.

Teams need practical guidance for reinforcement learning agents in AgentDB. This skill provides templates, training workflows, and evaluation patterns for learning plugins.

Advanced AgentDB projects need reliable patterns for sync, search, and database operations. This skill gives practical guidance for QUIC sync, hybrid search, sharding, and production tuning.

Teams need consistent agents that encode domain rules and workflows. This skill guides Claude, Codex, and Claude Code through agent design, SDK setup, testing, and packaging.

Distributed vector search needs careful replication, metrics, monitoring, and deployment planning. This skill gives a phased AgentDB workflow for scalable hybrid search systems.

Adding opinionated Oxlint rules manually requires copying plugin files and merging configuration safely. This skill performs the integration while preserving existing project settings.

Complex engineering work can suffer from shallow analysis and missed trade-offs. This skill helps Claude, Codex, and Claude Code escalate reasoning when complexity signals appear.

AI coding agents often write implementation code before proving expected behavior with tests. This skill enforces a red, green, refactor workflow for safer feature work, bug fixes, and refactoring.

Code reviews often lack context, which slows feedback and misses important risks. This skill guides structured review requests and response handling for Claude, Codex, and Claude Code.

Complex implementation plans can lose quality when tasks share stale context. This skill guides isolated subagent execution with review checkpoints after every task.

Large coding sessions often contain independent failures that slow one agent down. This skill helps Claude, Codex, and Claude Code split work safely across focused agents.

Ruby teams often need type signatures without changing source files. This skill guides Claude, Codex, and Claude Code to create accurate Sorbet RBI files and validate them with Sorbet.

Ruby teams often need typed method contracts without moving logic into RBI files. This skill guides Claude, Codex, and Claude Code through adding Sorbet sig blocks directly in source files.

Ruby teams need type signatures close to source code. This skill adds RBS-inline comments and guides validation without creating separate RBS files.

Ruby projects often lack complete RBS signatures, which limits static type checking. This skill guides Claude, Codex, or Claude Code through generating and validating pure RBS files from Ruby source.

Prisma v7 changed client generation, adapters, and configuration. This skill gives concise patterns for setup, migration, queries, and troubleshooting.

Code review documents can leave teams with a long, hard-to-track fix list. This skill turns those findings into an ordered workflow for fixes, skips, checks, and status updates.