Valstorm Hosted MCP Server
OfficialThis server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Valstorm Hosted MCP Server (apps/valstorm-mcp-server)
Official hosted Model Context Protocol (MCP) server for the Valstorm Enterprise Platform.
Official Tool Page & Quickstart: https://www.valstorm.com/free-tools/valstorm-mcp
All Valstorm Free & Open-Source Tools: https://www.valstorm.com/free-tools
Valstorm BOS Platform Overview: https://www.valstorm.com/bos
This microservice acts as a stateless, high-performance SSE & JSON-RPC proxy to the Valstorm Core API (https://api.valstorm.com/v1), exposing granularly scoped tools for autonomous agents (Claude Desktop, Cursor, hermes profiles, internal enterprise workers) while offloading binary transfer via direct S3 recipes.
Architecture Highlights
Dual Transport: Serves Streamable HTTP at
/mcp(the current MCP standard, used by Claude and Cursor), the legacy SSE transport at/sse+/messagesfor older clients, andstdiofor local CLI workflows.Stateless Transport:
/mcpruns stateless, so any replica can answer any request. Multi-replica deployments need no session affinity.OAuth 2.1 Discovery: Publishes RFC 9728 protected-resource and RFC 8414 authorization-server metadata describing itself, delegating authorization and token endpoints to the core API.
Stateless Proxy: Zero direct database attachments. Forwards JSON-RPC tool invocations to the core Valstorm REST API via
httpx.AsyncClient.Granular Scopes: Strict token scope inspection (
vfs:read,vfs:write,vfs:delete,record:read,record:write,record:delete,query:execute).Transfer Orchestrator: Bypasses JSON-RPC bloat by returning S3 presigned URLs and execution recipes for binary transfers.
RAG & Search: Server-side Gemini RAG (
vfs_ask) and OpenSearch hybrid retrieval (vfs_search).Zero N+1 Querying: Single-roundtrip document fetching (
vfs_read_textviaGET /vfs/file/{id}) and batch foreign key hydration (record_hydrate_batchviaPOST /object/hydrate-batch).
Related MCP server: Atlas MCP Server
Quickstart
Running locally with stdio:
uv run --project apps/valstorm-mcp-server valstorm-mcp --transport stdioRunning the hosted HTTP server (Streamable HTTP + SSE):
uv run --project apps/valstorm-mcp-server valstorm-mcp --transport streamable-http --host 0.0.0.0 --port 8660Or via uvicorn directly:
uv run --project apps/valstorm-mcp-server uvicorn valstorm_mcp.server:app --host 0.0.0.0 --port 8660Available Tool Registry (16 Tools)
Virtual File System (VFS)
vfs_search(query, limit=10, vault_id=None)— Hybrid OpenSearch vector & metadata chunk search.vfs_ask(query, vault_id=None, limit=5)— Server-side Gemini RAG with synthesized markdown answer and citations.vfs_read_text(file_id, max_bytes=100000)— High-performance single-roundtrip text reader (GET /vfs/file/{id}).vfs_browse(target='root', browse_type='vault')— Explores VFS vaults, folder trees, or full hierarchy snapshots.vfs_move(item_id, to_vault_id)— Atomic file or vault folder moves with path cascades.vfs_prepare_upload(filename, vault_id=None, content_type="application/octet-stream")— Generates out-of-band upload recipe.vfs_get_download_url(file_id)— Generates presigned S3 GET URL & terminal download recipe.vfs_delete_file(file_id, confirm=False)— Secure file deletion guarded by confirmation flag.
Records, SQL, & Schemas
record_query_sql(query, bypass_cache=False, limit=50)— Executes custom Valstorm SQL engine (SELECT ... WHERE owner = ME).record_query_mongo(collection, pipeline)— Structured document aggregation pipeline ($match,$group,$lookup,$facet).record_hydrate_batch(ids)— Resolves foreign key ID arrays to display names and schemas in a single roundtrip.record_get(api_name, record_id)— Fetches single record by ID.record_create(api_name, record)— Creates single or batch records.record_update(api_name, record_id, data)— Patches existing record fields.record_delete(api_name, record_id, confirm=False)— Removes a record guarded by confirmation flag.schema_inspect(api_name=None)— Introspects object schemas, fields, lookups, and validation rules.
License & Commercial Terms
Valstorm Hosted MCP Server is distributed under a Dual License model:
Noncommercial Use (Community & Academic): Licensed under the PolyForm Noncommercial License 1.0.0. You are free to run, inspect, and evaluate the server for personal, hobby, or educational purposes without fees.
Commercial & Enterprise Use: Any commercial use—including running internal business operations, powering SaaS backends, managed hosting, or client consulting deliverables—requires a commercial agreement.
For commercial licensing, enterprise BYOC terms, or official partner programs:
Email: team@valstorm.com
Website: Valstorm Pricing
Enterprise Portal: Valstorm Enterprise
See CONTRIBUTING.md for community guidelines and development policies.
Copyright (c) 2026 Valstorm LLC. All rights reserved.This server cannot be deployed
Maintenance
Related MCP Connectors
Your org's AI agents, tasks, runs, search, and brain files as MCP tools and resources.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
214 tools for AI agents over one MCP endpoint: image, video, audio, 3D, data, storage, mail, GPUs.
MCP-first toolbox for agents: KV storage, auth, queue, and utility tools. Free in early access.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables file system operations, web scraping, and AI-powered search through MCP tools for use by LLM agents.1-
- FlicenseNot gradedqualityDmaintenanceEnables AI agents and users to manage workspace files, monitor system metrics, take persistent notes, and retrieve weather data via MCP tools and resources.-
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to securely discover, invoke, and manage tools through a hardened MCP endpoint with protections like injection detection, circuit breakers, retry backoff, response caching, context-window limiting, and state snapshots.7 npmMIT
- AlicenseAqualityCmaintenanceEnables MCP-capable agents to create and invoke OpenServerless endpoints, manage secrets, and configure integrations such as S3, PostgreSQL, Redis, Milvus, and MongoDB.13Apache 2.0