MCP servers and skills: search, filter by where they run, and compare.
1–24 of 2,473 tools for “sonnet”
Clear all🧠 MCP server implementing RAT (Retrieval Augmented Thinking) - combines DeepSeek's reasoning with GPT-4/Claude/Mistral responses, maintaining conversation context between interactions.
SkillMIT★ 115No key
SkillMITFew starsNo key
Authenticated async Sonnet 5 Agent agent with status polling and artifact results.
Hosted by authorNo licenseFew starsNo key
LLM elevation protocol: make a daily-driver model (Sonnet) operate near frontier level by buying depth with inference compute — GEN×3, adversarial verification, external memory, and distilled-judgment
SkillMITFew starsNo key
SkillMITFew starsNo key
Claude Code skill: Opus plans and writes briefs, Sonnet workers execute - big tasks, fewer expensive tokens.
SkillMITFew starsNo key
Claude Code skill that stops Sonnet 5 wasting tokens: fewer rereads, fewer spawn-happy subagents, cheaper runs.
SkillLink onlyFew stars
Claude Code skill: equipo Sonnet(PM)/Opus(dev)/Haiku(QA+adversarial) con contrato de tests congelados
SkillLink only★ 1Few stars
Claude Code plugin: plan on your main model, code on Sonnet xhigh, review on Opus. The expensive model never writes app code.
SkillMIT★ 1Few starsNo key
Claude Code skill: route subagent workers by task profile. Frontier reasoning for planning, judging and synthesis; commodity pricing for execution. Based on Anthropic's benchmarked Fable-orchestrator/
SkillApache-2.0★ 1Few starsNo key
Dual-tier Claude API routing — Haiku fast, Sonnet deep. Cost-optimize LLM workloads without losing quality. pip package + Claude Code skill · MIT.
SkillMIT★ 1Few starsNo key
Claude Code plugin: model×effort worker agent grid + planner relay for Superpowers subagent dispatch
SkillMIT★ 1Few starsNo key
Create LangChain agents with create_agent, define tools, and use middleware for human-in-the-loop and error handling.
OfficialSkillMIT★ 1.3k in collectionNo key
Opus 指挥，干活模型执行，Haiku 验收，Sonnet 审查：靠流程约束，不靠模型自觉。
SkillMIT★ 1Few starsNo key
Determines which LLM model to use for each agent spawn
OfficialSkillMIT★ 5.4k in collectionNo key
Scaffold a minimal local LangGraph agent in TypeScript by following the official quickstart. Use when the user wants to quickly build or try a LangGraph agent locally.
OfficialSkillMIT★ 1.3k in collectionNo key
Scaffold a minimal local LangGraph agent in Python by following the official quickstart. Use when the user wants to quickly build or try a LangGraph agent locally.
OfficialSkillMIT★ 1.3k in collectionNo key
Claude plugin for rigorous biblical study. TDD-built to prevent AI failures: psychologizing texts, moralistic applications, genre mistakes, and exegetical shortcuts. Faithful, Christ-centered analysis
SkillGPL-3.0★ 7No key
LLM Pricing Oracle — MCP server with real pricing data from LiteLLM
Node.jsMIT★ 1Few starsNo key
Cost-aware model routing for Claude Code using Haiku, Sonnet, and Opus subagents, with failure tracking and experimental self-tuning.
SkillMITFew starsNo key
Stop paying the king to do clerk work. Cost-aware routing across Claude Fable, Opus, and Sonnet.
SkillMIT★ 1Few starsNo key
Automatic model routing for Claude Code — Fable 5, Opus 4.8, Sonnet 5, Haiku 4.5. Right model, right task, zero manual switching. Part of the VZT Protocol ecosystem.
SkillMIT★ 2Few starsNo key
Scaffold a minimal local LangChain agent in Python by following the official quickstart. Use when the user wants to quickly build or try a LangChain agent locally.
OfficialSkillMIT★ 1.3k in collectionNo key
Scaffold a minimal local LangChain agent in TypeScript by following the official quickstart. Use when the user wants to quickly build or try a LangChain agent locally.
OfficialSkillMIT★ 1.3k in collectionNo key
Who stands behind it
