Skip to content
MCP server

Contextator

Turn folders, repos, Notion and Confluence into one searchable MCP endpoint per project

Pricing
Open source
Platforms
self-hosted, Linux, API
Runs
Remotely
Version
1.0.0
Listing updated
3 days ago

About Contextator

Contextator is an open-source MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.

Contextator runs as a single Docker container that holds both PostgreSQL with pgvector and the app. Each project gets its own isolated MCP URL, such as /mcp/billing, built from any mix of sources: a mounted folder, a git repository, an upload, Notion, Confluence, a published documentation site, an S3 bucket, a Google Drive folder or a GitHub issue tracker. A client connected to one project never sees another.

Embeddings are computed locally on the CPU with a multilingual model by default, with OpenAI embeddings available through two environment variables. If the best hit scores below a configurable floor (0.82 by default), it replies with no good match rather than guessing. Files are hashed so only changed ones are re-embedded, and formats like PDF, Word, PowerPoint and Excel are converted to Markdown during indexing.

Access is by bearer token, open, or account-based, with roles for admins, editors and viewers. It is AGPL-3.0-or-later with a commercial licence offered. The project's own comparison admits it is weaker on setup effort and has no catalogue of public library docs.

What it does

One endpoint per project
Isolated collections and embeddings for each project.
Many source types
Folders, git, uploads, Notion, Confluence, doc sites, S3, Drive and GitHub issues.
Local embeddings
CPU-generated by default, so nothing leaves your machine with local sources.
No-guess threshold
Returns no good match below a score floor.
Incremental indexing
Only changed files are re-embedded.
Access controls
Bearer token, open or account-based, with roles and API tokens.

Good for

  • Give a coding agent a company handbook stored in Notion
  • Search internal docs across several repositories from one endpoint
  • Keep private documentation off hosted RAG services
  • Index PDFs and Word files alongside Markdown
  • Expose separate doc sets to different teams

Pricing

Open source

AGPL-3.0-or-later; commercial licence available.

Prices change. Check contextator.com before you buy.

Contextator FAQ

How do I run it?

As one Docker container, published as contextator/contextator, started with docker compose.

Is it free?

It is free software under AGPL-3.0-or-later, with a commercial licence available.

Does it send my documents elsewhere?

With local sources and a cached model there is no egress; OpenAI embeddings are optional.

Which public library docs are included?

None; it indexes only the sources you add.

Is Contextator free?

Yes. Contextator is open source, so you can use and self-host it for free.

What are the best alternatives to Contextator?

The closest alternatives to Contextator are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Contextator

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools