MidOS - MCP Community Library
Curated knowledge API for AI agents - skill packs, semantic search, validated patterns.
Ask AI about MidOS - MCP Community Library
Powered by Claude Β· Grounded in docs
I know everything about MidOS - MCP Community Library. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
MidOS β MCP Server for Developer Knowledge
Curated, validated knowledge for AI coding agents. Not raw docs β battle-tested patterns.
125 skill packs across 20+ tech stacks. 46,000+ curated chunks. 436 validated discoveries. Every piece reviewed, cross-validated, and myth-busted.
Your agent asks: "How do I implement optimistic updates in React 19?"
MidOS returns: Battle-tested pattern with useOptimistic + Server Actions, validated Feb 2026.
Context7 returns: Raw React docs from reactjs.org.
Quick Start
One line. Add to your MCP config and start querying:
Claude Code β .mcp.json or ~/.claude/settings.json
{
"mcpServers": {
"midos": {
"url": "https://midos.dev/mcp"
}
}
}
Cursor / Windsurf β MCP Settings
Add a new server:
- Name:
midos - URL:
https://midos.dev/mcp - Transport: Streamable HTTP
Cline β MCP Settings
{
"mcpServers": {
"midos": {
"url": "https://midos.dev/mcp",
"transportType": "streamable-http"
}
}
}
OpenCode β opencode.json
{
"mcp": {
"midos": {
"type": "remote",
"url": "https://midos.dev/mcp"
}
}
}
Or self-hosted (stdio):
{
"mcp": {
"midos": {
"type": "local",
"command": ["python", "-m", "modules.mcp_server.midos_mcp"],
"enabled": true
}
}
}
Self-hosted β Run locally
git clone https://github.com/MidOSresearch/midos-mcp.git
cd midos-mcp
pip install -e .
pip install -e hive_commons/
python -m modules.mcp_server.midos_mcp --http --port 8419
Then point your MCP client to http://localhost:8419/mcp.
First Tool Call
After connecting, personalize your experience:
agent_handshake(model="claude-opus-4-6", client="claude-code", languages="python,typescript", frameworks="fastapi,react")
Then search for what you need:
search_knowledge("React 19 Server Components patterns")
Tools Reference
Free Tier (no API key)
| Tool | Description | Example |
|---|---|---|
search_knowledge | Search 46,000+ curated chunks across all stacks | search_knowledge("FastAPI dependency injection") |
hybrid_search | Combined keyword + semantic search with reranking | hybrid_search("PostgreSQL JSONB indexing") |
list_skills | Browse 125 skill packs by technology | list_skills(stack="react") |
get_skill | Get a specific skill pack (preview in free, full in Dev) | get_skill("nextjs") |
get_protocol | Protocol and pattern documentation | get_protocol("domain-driven-design") |
hive_status | System health and live statistics | hive_status() |
project_status | Knowledge pipeline dashboard | project_status() |
agent_handshake | Personalized onboarding for your model + stack | See example above |
Dev Tier ($20/mo β full content + advanced search)
| Tool | Description | Example |
|---|---|---|
get_eureka | Validated breakthrough discoveries (436 items) | get_eureka("response-cache") |
get_truth | Empirically verified truth patches (52 items) | get_truth("qlora-myths") |
semantic_search | Vector search with Gemini embeddings (3072-d) | semantic_search("event sourcing CQRS") |
research_youtube | Extract knowledge from video content | research_youtube("https://youtube.com/...") |
chunk_code | Intelligent code chunking for ingestion | chunk_code(code="...", language="python") |
memory_stats | Vector store analytics and health | memory_stats() |
episodic_search | Search agent session history | episodic_search("last deployment issue") |
Ops Tier (custom β security, infrastructure, advanced ops)
Contact for specialized knowledge packs. midos.dev/pricing
Skill Packs (125 and growing)
Production-tested patterns for:
Frontend: React 19, Next.js 16, Angular 21, Svelte 5, Tailwind CSS v4, Remix v2
Backend: FastAPI, Django 5, NestJS 11, Laravel 12, Spring Boot, Symfony 8
Languages: TypeScript, Go, Rust, Python
Data: PostgreSQL, Redis, MongoDB, Elasticsearch, LanceDB, Drizzle ORM, Prisma 7
Infrastructure: Kubernetes, Terraform, Docker, GitHub Actions
AI/ML: LoRA/QLoRA, MCP patterns, multi-agent orchestration, Vercel AI SDK
Testing: Playwright, Vitest
Architecture: DDD, GraphQL, event-driven, microservices, spec-driven dev
How MidOS is Different
| Raw Docs (Context7, etc.) | MidOS | |
|---|---|---|
| Content | Documentation dumps | Curated, human-reviewed, cross-validated |
| Quality | No validation | 5-layer pipeline: chunks β truth β EUREKA β SOTA |
| Search | Keyword matching | Semantic + hybrid search (Gemini embeddings, 3072-d) |
| Onboarding | Generic | Personalized per model + CLI + stack |
| Format | Raw text | Stack-specific skill packs with production patterns |
| Accuracy | Stale docs | Myth-busted with empirical evidence |
Knowledge Pipeline
staging/ β chunks/ β skills/ β truth/ β EUREKA/ β SOTA/
(entry) (L1) (L2) (L3) (L4) (L5)
- Chunks (46,000+): Curated, indexed knowledge across 20+ stacks
- Skills (125): Organized, actionable, versioned by stack
- Truth (52): Verified with empirical evidence
- EUREKA (436): Validated improvements with measured ROI
- SOTA (138): Best-in-class, currently unimprovable
Using an API Key
Pass your key via the Authorization header for Dev/Ops access:
{
"mcpServers": {
"midos": {
"url": "https://midos.dev/mcp",
"headers": {
"Authorization": "Bearer midos_your_key_here"
}
}
}
}
Get a key at midos.dev/pricing.
Architecture
midos-mcp/
βββ modules/mcp_server/ FastMCP server (streamable-http)
βββ knowledge/
β βββ chunks/ Curated knowledge (L1) β 46,000+ items
β βββ skills/ Stack-specific skill packs (L2) β 125 items
β βββ EUREKA/ Validated discoveries (L4) β 436 items
β βββ truth/ Empirical patches (L3) β 52 items
βββ hive_commons/ Shared library (LanceDB vector store, config)
βββ smithery.yaml Smithery marketplace manifest
βββ Dockerfile Production container
βββ pyproject.toml Dependencies and build config
Tech Stack
- Server: FastMCP 2.x (streamable-http transport)
- Vectors: LanceDB + Gemini embeddings (670,000+ vectors, 3072-d)
- Auth: 2-tier API key middleware (free β dev) with rate limiting
- Pipeline: 5-layer quality validation with myth-busting
- Deploy: Docker + Coolify (auto-deploy on push)
Contributing
MidOS is community-first. If you have production-tested patterns, battle scars, or discovered that a popular claim is false β we want it.
- Search existing knowledge first:
search_knowledge("your topic") - Open an issue describing the pattern or discovery
- We'll review and add it to the pipeline
License
Source-verified developer knowledge. Built by devs, for agents.
midos.dev Β· Discussions Β· Issues
