Hosted MCP server · #94 most used on Smithery

Quelvio

quelvio/mcp-server

**Your company's brain for AI agents.** Quelvio continuously understands your organization's knowledge across every connected system — what's authoritative, what's recent, what's persistent, what's temporary, and what should be forgotten. Returns cited answers grounded in your authoritative sources, refusing when the corpus doesn't actually answer the question. ## What makes Quelvio different - **Cross-document supersession** — when sources conflict, surfaces the current truth and explicitly names what it supersedes - **Authority-weighted retrieval** — weights documents by who said them on the relevant topic - **Refuses on weak context** — when the corpus doesn't actually answer, says so. No hallucinated numbers, no fabricated citations - **Cross-source synthesis** — reasons across Slack, Confluence, Notion, Drive simultaneously with explicit source attribution - **Lifecycle awareness** — tracks current vs deprecated content, flags stale procedures ## Tools - `query_knowledge` — search across all connected sources with optional synthesis modes (fast / standard / deep) - `list_domains` — discover what knowledge domains your tenant has indexed - `get_source_detail` — chunk-level provenance for any cited result ## Setup 1. Sign up at https://quelvio.com — 5GB indexed + 2 queries/day free tier 2. Connect your sources (Google Drive, SharePoint, Confluence, Slack, Notion, GitHub) 3. Add this MCP server to your client 4. Complete OAuth on first tool call ## Authentication OAuth 2.1 with PKCE. Per-employee identity propagation — each query is scoped to the individual employee's permissions in source systems. ## Pricing Per Knowledge Token model. Paid plans scale with usage and seats.

Microsoft Office & docs Security Git & GitHub Memory & RAG Docs & writing Productivity & PM

Install ↓Smithery ↗Website ↗

Uses: 5.9k
Tools: 3
Runs: Hosted
Listed: 6mo ago

Get started

Add it to your agent

npx -y @smithery/cli@latest install quelvio/mcp-server --client claude

Swap --client claude for cursor, vscode, windsurf or another client. The CLI asks for any required settings.

claude mcp add --transport http quelvio https://server.smithery.ai/quelvio/mcp-server/mcp

A hosted server: nothing to install. Run /mcp in Claude Code to sign in if it asks.

[mcp_servers.quelvio]
url = "https://server.smithery.ai/quelvio/mcp-server/mcp"

Add to ~/.codex/config.toml.

{
  "mcpServers": {
    "quelvio": {
      "url": "https://server.smithery.ai/quelvio/mcp-server/mcp"
    }
  }
}

Add to ~/.cursor/mcp.json.

{
  "servers": {
    "quelvio": {
      "type": "http",
      "url": "https://server.smithery.ai/quelvio/mcp-server/mcp"
    }
  }
}

Add to .vscode/mcp.json.

gemini mcp add --transport http quelvio https://server.smithery.ai/quelvio/mcp-server/mcp

Saved to .gemini/settings.json.

What it can do

3 tools

query_knowledge4 paramsSearch the company's connected knowledge across every source — Drive, SharePoint, Confluence, Slack, Notion — with cited synthesized answ…

Search the company's connected knowledge across every source — Drive, SharePoint, Confluence, Slack, Notion — with cited synthesized answers, lifecycle awareness, and refusal-on-weak-context. Returns a written answer with [n] citations plus the ranked source chunks. Modes: `fast` (1,500 kT — retrieval-only, no synthesis), `standard` (12,500 kT — default; synthesized answer over the top retrieval set), `deep` (25,000 kT — wider retrieval + premium synthesis for complex questions). Pick the cheape

Parameter	Type	Description
`mode`	string	fast \| standard \| deep. Defaults to `standard`. `fast` returns chunks only (no synthesized answer); `standard` and `deep` return a synthesized answer with [n] citations + the source chunks.
`query`	string	Natural-language query (1–2000 characters). Be specific — results are ranked by authority + relevance, not keyword overlap.
`domain`	string	Optional taxonomy domain filter (e.g. 'engineering.platform'). Use `list_domains` to discover valid values for the tenant.
`max_sources`	number	Number of source chunks to return (1–20, default 5). The 25K token cap may force fewer results regardless of this value.

list_domains1 paramList the taxonomy domains the company has indexed — with document counts, expert counts, and coverage levels — so an agent can decide whe…

List the taxonomy domains the company has indexed — with document counts, expert counts, and coverage levels — so an agent can decide whether to query before spending a Knowledge Token. Returns one row per domain with the canonical `taxonomy_domain` slug, document/chunk counts, expert count, coverage level (expert | partial | none), the single_expert risk flag, and the top contributor by authority. Use the slug as the `domain` filter on a follow-up `query_knowledge` call. Zero Knowledge Tokens c

Parameter	Type	Description
`coverage_filter`	string	Optional comma-separated subset of expert,partial,none. Default: all three. Unknown tokens 400.

get_source_detail1 paramReturn per-chunk source provenance for a previous query — document path, lifecycle state, embedding timestamp, contributor, last-updated …

Return per-chunk source provenance for a previous query — document path, lifecycle state, embedding timestamp, contributor, last-updated — useful for verifying a citation or surfacing trust signals to a downstream system. Pass a `query_id` returned by an earlier `query_knowledge` call. Returns 404 if the query_id is unknown OR belongs to a different tenant (indistinguishable to prevent info-leak). Zero Knowledge Tokens consumed.

Parameter	Type	Description
`query_id`	string	UUID returned in the structured-metadata block of a prior `query_knowledge` response. Tenant-scoped — cross-tenant 404.