MCP App StoreMCP App Store
from ai-toolkit155
Guidelines and patterns for implementing prompt caching with Anthropic APIs to reduce input-token costs and latency, including TTLs, breakpoints, and hit-rate m
This skill documents practical patterns for implementing prompt caching when building with Anthropic (Claude) APIs. It explains cache mechanics, recommended TTLs, how to structure cached blocks and breakpoints, anti-patterns that cause cache misses, and methods for measuring cache hit rate. Concrete code examples in Python and TypeScript show how to mark cached content and keep dynamic data outside cached prefixes.
Use this skill when you need to dramatically cut API input-token costs or reduce latency by caching stable parts of prompts. It's useful during production integration, performance tuning, or any high-volume loop where repeated stable prefixes exist (system prompts, tool definitions, large reference docs). Avoid for one-shot calls or very small prompts.
Likely used by agents and tooling that call Anthropic/Claude APIs (Claude-opus/Haiku), and by developer-facing tools (Claude Code, Cursor, Copilot integrations) that need token-cost optimizations.
Pure knowledge/documentation skill with no scripts. Covers Anthropic prompt caching mechanics, layering patterns, anti-patterns, and hit-rate measurement with code examples in Python and TypeScript. Well-structured reference content — clear tables, practical guidance, and specific threshold numbers (0.7 hit-rate target, 1024/2048 min tokens). No security concerns whatsoever.
Clean knowledge-only skill. Would benefit from a references/ directory linking to official Anthropic docs and related skills. The skill_md_body already references the docs URL inline which is good. User-invocable is false, so it's designed as a passive reference skill triggered by keywords like 'prompt caching' and 'cache_control'.
Ubiquitous Language Glossary
Extract a DDD-style ubiquitous language glossary from conversation: identify domain terms, flag ambiguities, propose canonical terms, and write the result to UB
Memory Search
Search past coding sessions and observations using natural-language queries to retrieve decisions, notes, and context.
Documentation Generator
Generate and update README, API docs, architecture notes, changelogs and knowledge-base entries from a codebase, with templates and review guidance.
Design Engineering
Guidelines and patterns for UI animation, micro-interactions, and component polish to build performant, accessible, and delightful interfaces.
Kotlin Rules
Project-specific Kotlin coding standards covering naming, null-safety, coroutines, testing, frameworks (Ktor, Spring), and security best practices.
Yielding Bear — Unified LLM Routing
A unified LLM routing API that automatically routes requests to the cheapest capable model across providers to reduce inference cost by ~60-80%.
Caveman Compress
Compress natural-language memory files into a compact 'caveman' format to save LLM tokens while preserving code, links, and structure.
FastAPI Project Templates
Creates production-ready FastAPI project scaffolds with async patterns, DI, middleware, and testing best practices for high-performance APIs.
GraphQL Inspector — Validate
Validate GraphQL operations against a schema with configurable depth, complexity, alias and directive limits to catch errors before runtime.
OpenTestAI
Automated, high-confidence AI testing: bug detection, persona feedback, and prioritized test-case generation using many specialized tester profiles.
JJ Hunk
Programmatic hunk selection for Jujutsu (jj): list, split, commit, or squash specific hunks to enable partial commits and fine-grained history editing without a
Codex Collab
Bridge Claude and Codex to run tasks, perform code reviews, and run background jobs for research or implementation workflows.
Cursor Hello World Tutorial
A concise 10-minute hands-on tutorial demonstrating Cursor's Tab completion, Chat, Inline Edit, and Composer features for getting started quickly.