Routing specificity 25% weight
How clearly intents map to named components.
Sort any column and filter the complete research set. Criterion scores are analyst routing heuristics from documented scope, setup, maturity, local control, security clarity, composability, and documentation—not popularity or intrinsic quality.
| Repository | Primary type | Domains | Maturity | Setup effort | Production readiness | Routing specificity | Setup simplicity | Operational maturity | Composability | Local control | Security boundary clarity | First-party routing documentation | Weighted % |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| amaancoderx/npxskillui | CLI | frontend design, design systems | established | low | development | 4 | 5 | 3 | 4 | 5 | 3 | 5 | 82.0 |
| anthropics/claude-code | CLI | software engineering, AI coding agents | established | low | production | 4 | 5 | 5 | 4 | 5 | 3 | 5 | 88.0 |
| anthropics/skills | Skill collection | agent skills, documents, design, development | established | low | production | 4 | 5 | 5 | 4 | 5 | 3 | 5 | 88.0 |
| coleam00/hyperframes-ai-video-generation | Workflow | AI video, short-form content | established | high | development | 4 | 1 | 3 | 4 | 3 | 3 | 5 | 70.0 |
| D4Vinci/Scrapling | Library | web scraping, browser automation, data acquisition | established | medium | production | 4 | 3 | 5 | 4 | 5 | 5 | 5 | 88.0 |
| diegosouzapw/OmniRoute | API | LLM routing, token compression, agent infrastructure | established | medium | development | 4 | 3 | 3 | 4 | 3 | 5 | 5 | 78.0 |
| firecrawl/firecrawl | API | web scraping, web search, agent data acquisition | established | low | production | 4 | 5 | 5 | 4 | 3 | 5 | 5 | 88.0 |
| Leonxlnx/taste-skill | Skill collection | frontend design, visual design | established | low | development | 4 | 5 | 3 | 4 | 5 | 3 | 5 | 82.0 |
| n8n-io/n8n | Framework | automation, AI agents, integrations | established | medium | production | 4 | 3 | 5 | 4 | 3 | 5 | 5 | 84.0 |
| nextlevelbuilder/ui-ux-pro-max-skill | Skill collection | UI/UX, frontend design | established | low | development | 4 | 5 | 3 | 4 | 5 | 3 | 5 | 82.0 |
| nexu-io/open-design | Standalone application | design, prototyping, creative artifacts | established | medium | development | 4 | 3 | 3 | 4 | 5 | 3 | 5 | 78.0 |
| Panniantong/Agent-Reach | CLI | agent web access, social platforms, research | established | medium | development | 4 | 3 | 3 | 4 | 5 | 5 | 5 | 82.0 |
| pbakaus/impeccable | Skill collection | frontend design, UX, UI quality | established | low | production | 4 | 5 | 5 | 4 | 5 | 3 | 5 | 88.0 |
| supabase/supabase | Monorepo | database, backend, auth, storage, realtime | established | medium | production | 4 | 3 | 5 | 4 | 5 | 5 | 5 | 88.0 |
| vercel/vercel | Monorepo | web deployment, frontend infrastructure | established | low | production | 3 | 5 | 5 | 4 | 3 | 3 | 5 | 79.0 |
| jarrodwatts/claude-hud | Plugin | Claude Code, observability | established | low | development | 3 | 5 | 3 | 4 | 5 | 3 | 5 | 77.0 |
| virgiliojr94/book-to-skill | Skill collection | agent skills, knowledge management, documents | established | medium | development | 4 | 3 | 3 | 4 | 5 | 3 | 5 | 78.0 |
| github/spec-kit | CLI | software planning, spec-driven development | established | medium | production | 4 | 3 | 5 | 4 | 5 | 3 | 5 | 84.0 |
| gsd-build/get-shit-done | Documentation repository | software planning | deprecated | low | unknown | 3 | 5 | 0 | 5 | 5 | 3 | 5 | 71.0 |
| bmad-code-org/BMAD-METHOD | Framework | software engineering, agile AI, multi-agent planning | established | medium | production | 4 | 3 | 5 | 4 | 5 | 3 | 5 | 84.0 |
| mattpocock/skills | Skill collection | software engineering, agent skills | established | low | production | 4 | 5 | 5 | 4 | 5 | 3 | 5 | 88.0 |
| Agent-Field/SWE-AF | Framework | autonomous software engineering, multi-agent orchestration | emerging | high | development | 4 | 1 | 3 | 4 | 3 | 5 | 5 | 74.0 |
| krusemediallc/arcads-claude-code | Skill collection | advertising, AI video, AI images | established | medium | development | 4 | 3 | 3 | 4 | 3 | 3 | 5 | 74.0 |
| thedotmack/claude-mem | Plugin | Claude Code, agent memory, context engineering | established | medium | production | 4 | 3 | 5 | 4 | 5 | 5 | 5 | 88.0 |
| langgptai/awesome-claude-prompts | Prompt collection | prompt engineering, Claude | established | low | prototype | 3 | 5 | 1 | 4 | 5 | 3 | 5 | 71.0 |
| hesreallyhim/awesome-claude-code | Documentation repository | Claude Code, developer tools | established | low | production | 3 | 5 | 5 | 5 | 5 | 3 | 5 | 86.0 |
| punkpeye/awesome-mcp-servers | Documentation repository | MCP, agent tools | unknown | low | unknown | 3 | 5 | 1 | 4 | 5 | 3 | 5 | 71.0 |
| zilliztech/claude-context | MCP server | semantic code search, codebase context | established | medium | production | 3 | 3 | 5 | 4 | 3 | 3 | 5 | 75.0 |
| lcoutodemos/clui-cc | Standalone application | Claude Code, desktop UI | established | high | development | 4 | 1 | 3 | 4 | 5 | 5 | 5 | 78.0 |
| multica-ai/andrej-karpathy-skills | Skill collection | coding-agent behavior, software engineering | established | low | production | 3 | 5 | 5 | 4 | 5 | 3 | 5 | 83.0 |
| shuairongzeng/claudia | Standalone application | Claude Code, desktop UI, agent management | established | high | development | 4 | 1 | 3 | 4 | 5 | 3 | 5 | 74.0 |
| openai/codex-plugin-cc | Plugin | Claude Code, Codex, code review | established | low | production | 4 | 5 | 5 | 4 | 3 | 3 | 5 | 84.0 |
| ComposioHQ/awesome-claude-skills | Skill collection | agent skills, MCP, integrations | established | low | production | 3 | 5 | 5 | 4 | 3 | 5 | 5 | 83.0 |
| EveryInc/compound-engineering-plugin | Skill collection | software engineering, knowledge compounding | established | medium | production | 4 | 3 | 5 | 4 | 5 | 3 | 5 | 84.0 |
| DenchHQ/DenchClaw | Standalone application | CRM, sales operations | established | medium | development | 3 | 3 | 3 | 4 | 3 | 3 | 5 | 69.0 |
| black-forest-labs/flux2 | Library | image generation, image editing, generative AI | established | high | production | 4 | 1 | 5 | 4 | 3 | 5 | 5 | 80.0 |
| florykhan/FutureSpend | Standalone application | personal finance, calendar, AI coaching | experimental | medium | prototype | 4 | 3 | 1 | 4 | 5 | 5 | 5 | 76.0 |
| darnoux/claude-code-level-up | Skill collection | Claude Code, developer education | emerging | low | development | 3 | 5 | 3 | 4 | 5 | 3 | 5 | 77.0 |
| hyperlight-dev/hyperagent | Framework | AI agents, sandboxing, secure automation | experimental | high | prototype | 4 | 1 | 1 | 4 | 5 | 5 | 5 | 72.0 |
| karpathy/autoresearch | Framework | LLM research, autonomous experimentation | experimental | medium | prototype | 3 | 3 | 1 | 4 | 5 | 5 | 5 | 71.0 |
| karpathy/nanochat | Framework | LLM training, ML research | established | medium | development | 4 | 3 | 3 | 4 | 5 | 3 | 5 | 78.0 |
| modelcontextprotocol/servers | MCP server | MCP, agent tools, protocol development | established | low | prototype | 4 | 5 | 1 | 4 | 5 | 5 | 5 | 80.0 |
| multica-ai/multica | Standalone application | coding agents, team collaboration, orchestration | established | high | production | 4 | 1 | 5 | 4 | 5 | 5 | 5 | 84.0 |
| VectifyAI/PageIndex | Library | RAG, document retrieval, long documents | established | medium | development | 4 | 3 | 3 | 4 | 3 | 3 | 5 | 74.0 |
| yamadashy/repomix | CLI | codebase context, LLM context | established | low | production | 4 | 5 | 5 | 4 | 5 | 5 | 5 | 92.0 |
| rtk-ai/rtk | CLI | token efficiency, developer tools, coding agents | established | low | production | 4 | 5 | 5 | 4 | 5 | 5 | 5 | 92.0 |
| obra/superpowers | Skill collection | software engineering, AI coding agents, TDD | established | low | production | 4 | 5 | 5 | 4 | 5 | 3 | 5 | 88.0 |
| VoltAgent/awesome-claude-code-subagents | Agent collection | Claude Code, subagents, software engineering | established | low | production | 4 | 5 | 5 | 4 | 5 | 3 | 5 | 88.0 |
| kepano/obsidian-skills | Skill collection | Obsidian, knowledge management, agent skills | established | low | production | 5 | 5 | 5 | 5 | 5 | 4 | 5 | 94.0 |
| JuliusBrussee/caveman | CLI | token efficiency, coding agents, context management | established | medium | development | 5 | 4 | 4 | 4 | 5 | 4 | 5 | 88.0 |
| DeusData/codebase-memory-mcp | MCP server | code intelligence, MCP, knowledge graphs | established | low | production | 5 | 5 | 5 | 5 | 5 | 5 | 5 | 100.0 |
| blader/humanizer | Skill collection | writing, editing, agent skills | established | low | production | 5 | 5 | 5 | 5 | 5 | 5 | 5 | 100.0 |
| coreyhaines31/marketingskills | Skill collection | marketing, growth, agent skills | established | low | production | 5 | 5 | 5 | 5 | 5 | 4 | 5 | 94.0 |
| ComposioHQ/composio | Monorepo | agent integrations, authentication, tool execution | established | medium | production | 5 | 4 | 5 | 5 | 3 | 4 | 5 | 86.0 |
| calesthio/OpenMontage | Workflow | AI video, video production, creative agents | emerging | high | development | 5 | 2 | 3 | 4 | 5 | 3 | 5 | 76.0 |
| Yeachan-Heo/oh-my-claudecode | Plugin | Claude Code, multi-agent orchestration, software engineering | established | medium | development | 5 | 4 | 4 | 3 | 4 | 3 | 5 | 80.0 |
How clearly intents map to named components.
Lower setup/operational burden scores higher.
Production-oriented evidence; experimental warnings reduce score.
Ability to coexist without conflicting workflow ownership.
Ability to run/self-host/control execution and data locally.
Explicitness of permissions, data boundaries, and limitations.
Explicit setup, component and limitation documentation.