What is Provenote and how does this MCP server integrate with it?

Provenote is a platform for managing research, drafts, and auditable runs. This MCP server is the official integration that lets you search your knowledge base, manage research threads, and track auditable runs directly from your AI chat interface.

How do I authenticate with Provenote through this MCP server?

Check the server's environment variable configuration in your MCP settings. You'll need valid Provenote credentials or an API key to connect. Refer to the official Provenote documentation for authentication setup details.

Can I search across all my Provenote knowledge and drafts at once?

Yes, the server supports knowledge search across your Provenote workspace. You can query drafts, research threads, and auditable runs as separate resources or search broadly depending on your query.

What does 'auditable runs' mean in Provenote?

Auditable runs are recorded executions or analyses in Provenote that maintain a full history and trace of inputs, outputs, and parameters. This MCP server lets you retrieve and review these runs to ensure reproducibility and accountability.

Is this server open-source and can I self-host it?

Yes, this is an open-source MCP server maintained at io.github.xiaojiou176-open/provenote-mcp. You can review the source code and self-host if needed, though the first-party Provenote integration is recommended for official support.

MCP ServerUnknownOfficialv1.8.5

Provenote MCP Server

Search and manage Provenote drafts, research threads, and auditable runs with full knowledge base integration. Built for developers who need traceable, searchable research workflows within their AI tools.

io.github.xiaojiou176-open/provenote-mcp

Open in Agent Studio Scan security View source

Hosted URL

Not published

Transport

Unknown

Auth

No auth required

What the Provenote MCP server does

How models use it and what it is built for.

Resources

Where to find authoritative docs and source for Provenote.

Example prompts for Provenote

Paste any of these into Agent Studio after connecting Provenote.

Search my knowledge base for papers on transformer architectures
Create a new research thread for my current project analysis
Show me all auditable runs from the past week with their outputs
What drafts do I have in progress and what's their current status

Provenote MCP server — FAQ

Common questions about connecting and running Provenote.

What is Provenote and how does this MCP server integrate with it?
Provenote is a platform for managing research, drafts, and auditable runs. This MCP server is the official integration that lets you search your knowledge base, manage research threads, and track auditable runs directly from your AI chat interface.
How do I authenticate with Provenote through this MCP server?
Check the server's environment variable configuration in your MCP settings. You'll need valid Provenote credentials or an API key to connect. Refer to the official Provenote documentation for authentication setup details.
Can I search across all my Provenote knowledge and drafts at once?
Yes, the server supports knowledge search across your Provenote workspace. You can query drafts, research threads, and auditable runs as separate resources or search broadly depending on your query.
What does 'auditable runs' mean in Provenote?
Auditable runs are recorded executions or analyses in Provenote that maintain a full history and trace of inputs, outputs, and parameters. This MCP server lets you retrieve and review these runs to ensure reproducibility and accountability.
Is this server open-source and can I self-host it?
Yes, this is an open-source MCP server maintained at io.github.xiaojiou176-open/provenote-mcp. You can review the source code and self-host if needed, though the first-party Provenote integration is recommended for official support.

Run Provenote across 60+ AI models, side-by-side

Connect Provenote to Claude, GPT, Gemini, DeepSeek and 60+ AI models in MCP Agent Studio. Compare answers side-by-side, save reusable agent presets, share runs — all in your browser, no install required.

Open Agent Studio