Semantic long-term memory storage and search using Qdrant vector database
npx claudepluginhub joshuacastanedaah/sean-mcp-servers --plugin qdrant-memoryA Claude Code plugin marketplace with standalone MCP servers for Microsoft 365, Azure DevOps, Dataverse, and Qdrant. Each plugin includes MCP tools + skills that teach Claude how to use them effectively.
99 tools across 6 plugins, 9 skills included.
Add this repo as a Claude Code plugin marketplace:
claude plugins marketplace add https://github.com/JoshuaCastanedaAH/sean-mcp-servers.git
Then install individual plugins:
claude plugins install microsoft-teams
claude plugins install azure-devops
# etc.
| Plugin | Tools | Skills | Description |
|---|---|---|---|
| microsoft-teams | 21 | 1 | Teams — chats, channels, meetings, polls, presence |
| microsoft-outlook | 15 | 1 | Outlook — send, read, search, drafts, folders |
| microsoft-sharepoint | 12 | 1 | SharePoint & OneDrive — files, sites, attachments |
| azure-devops | 35 | 4 | DevOps — work items, repos, PRs, pipelines, wikis |
| microsoft-dataverse | 14 | 1 | Dataverse — entities, agents, connectors |
| qdrant-memory | 2 | 1 | Qdrant vector memory — semantic store and search |
Each plugin is a self-contained Claude Code plugin with:
plugins/<name>/
.claude-plugin/plugin.json # Plugin metadata
.mcp.json # MCP server configuration
skills/<name>/SKILL.md # Skills teaching Claude how to use the tools
server/ # MCP server source (TypeScript)
src/index.ts # Server entry point + tool definitions
src/auth.ts # Authentication helpers
package.json
tsconfig.json
README.md # Server documentation
Skills are what set these apart from raw MCP servers — they include best practices, workflows, common patterns, and domain knowledge so Claude knows when and how to use each tool effectively.
If you prefer to use individual servers without the plugin system:
cd plugins/microsoft-teams/server
npm install
npm run build
.mcp.json{
"mcpServers": {
"teams": {
"command": "node",
"args": ["/path/to/plugins/microsoft-teams/server/dist/index.js"],
"env": {
"GRAPH_ACCESS_TOKEN": "your-token",
"GRAPH_REFRESH_TOKEN": "your-refresh-token",
"GRAPH_CLIENT_ID": "your-client-id",
"GRAPH_TENANT_ID": "your-tenant-id"
}
}
}
}
claude mcp add teams node /path/to/plugins/microsoft-teams/server/dist/index.js \
-e GRAPH_ACCESS_TOKEN=your-token
Each server uses environment variables for auth. No interactive login flows.
| Env Var | Required | Description |
|---|---|---|
GRAPH_ACCESS_TOKEN | Yes | OAuth 2.0 access token with Graph API scopes |
GRAPH_REFRESH_TOKEN | No | Refresh token for automatic token renewal |
GRAPH_CLIENT_ID | No* | Azure AD app registration client ID |
GRAPH_TENANT_ID | No* | Azure AD tenant ID |
* Required if using GRAPH_REFRESH_TOKEN.
Required Graph API permissions:
| Env Var | Required | Description |
|---|---|---|
DEVOPS_PAT | Yes | Personal Access Token |
DEVOPS_ORG_URL | Yes | Organization URL (e.g. https://dev.azure.com/myorg) |
| Env Var | Required | Description |
|---|---|---|
DATAVERSE_CLIENT_ID | Yes | Azure AD app client ID |
DATAVERSE_CLIENT_SECRET | Yes | Client secret |
DATAVERSE_TENANT_ID | Yes | Azure AD tenant ID |
DATAVERSE_ENVIRONMENT_URL | Yes | Environment URL (e.g. https://myorg.crm.dynamics.com) |
| Env Var | Required | Description |
|---|---|---|
QDRANT_URL | No | Qdrant instance URL (default: http://localhost:6333) |
QDRANT_API_KEY | No | Qdrant API key (if using Qdrant Cloud) |
EMBEDDING_API_URL | No | Embedding endpoint (default: http://localhost:3000/api/embeddings) |
cd plugins/<name>/server
npm install
npm run dev # Hot-reload with tsx
npm run build # Compile TypeScript
npm start # Run compiled version
MIT
Graph-vector memory service providing AI assistants with durable, relational memory for context awareness.
Admin access level
Server config contains admin-level keywords
Requires secrets
Needs API keys or credentials to function
Share bugs, ideas, or general feedback.
Pinecone vector database integration. Streamline your Pinecone development with powerful tools for managing vector indexes, querying data, and rapid prototyping. Use slash commands like /quickstart to generate AGENTS.md files and initialize Python projects and /query to quickly explore indexes. Access the Pinecone MCP server for creating, describing, upserting and querying indexes with Claude. Perfect for developers building semantic search, RAG applications, recommendation systems, and other vector-based applications with Pinecone.
Persistent memory for Claude Code — memories survive across sessions, projects, and machines
Open-source, local-first Claude Code plugin for token reduction, context compression, and cost optimization using hybrid RAG retrieval (BM25 + vector search), reranking, AST-aware chunking, and compact context packets.
Write SQL, explore datasets, and generate insights faster. Build visualizations and dashboards, and turn raw data into clear stories for stakeholders.
Agent Skills for AI/ML tasks including dataset creation, model training, evaluation, and research paper publishing on Hugging Face Hub