Skip to content
MCP server

Skeletable

Read and build source-backed tables where every value cites its origin

Install
Pricing
Open source
Platforms
Web, API
Runs
Remotely
Version
1.0.0
Listing updated
3 days ago

About Skeletable

Skeletable is an open-source MCP server for analytics and data. It runs as a hosted remote server, so there is nothing to install locally.

Each cell in a Skeletable table records where its value came from: a linked page, a person, a file or another app, or a model. Tables in the public library require proof, so a value links to the source and the exact words that state it, or says where the author looked. Reading needs no account.

The server's check_claim tool tests a statement against a cell and answers that it matches, differs, is disputed or is unknown. Agents can also create private drafts, add up to 500 rows per call, fill cells with proof and propose corrections that the table owner reviews. An agent can only write to drafts its person runs, every value still passes the checks, and a person publishes.

Datasets with optional proof can have AI columns with an instruction, such as classifying a pitch. Your own agent answers them through next_ai_cells and fill_ai_cells, so Skeletable never runs a model itself. The source is Apache-2.0 and published table data is under CC BY 4.0.

Install

Run in a terminal

claude mcp add --transport http skeletable https://skeletable.com/mcp

Add to Settings → Connectors → Add custom connector

https://skeletable.com/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "skeletable": {
      "url": "https://skeletable.com/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "skeletable": {
      "type": "http",
      "url": "https://skeletable.com/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "skeletable": {
      "serverUrl": "https://skeletable.com/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "skeletable": {
      "type": "http",
      "url": "https://skeletable.com/mcp"
    }
  }
}

Tools this server provides (16)

Taken from the project documentation; the running server may differ.

list_tables
Find a table by what it covers.
get_table
Read a table's scope, columns, column summaries and rows.
get_cell
Return one value's source, exact quote and date.
check_claim
Test a statement against a cell.
export_table
Export a table as CSV, optionally with sources.
propose_change
Suggest a new value with proof for the owner to review.
create_table
Make a private draft table.
add_columns
Add columns to a draft, including AI columns.
add_rows
Add up to 500 rows per call.
fill_cells
Fill a draft's cells with proof.
data_link
Get a private CSV address for a table.
save_version
Freeze a table as a numbered version.
fork_table
Copy a table into your account as a private draft.
set_instruction
Set or clear an AI column's instruction.
next_ai_cells
Get the next prompts for AI columns.
fill_ai_cells
Save an agent's answers to an AI column.

What it does

Cell-level provenance
Each value stores its source, exact quote and date, or where the author looked.
Claim checking
check_claim reports whether a statement matches, differs, is disputed or unknown.
Reviewable corrections
propose_change sends a new value with proof to the table owner.
Agent-built drafts
Create tables, add columns and rows, and fill cells in private drafts.
AI columns
Your own agent answers per-row instructions; answers record the model and prompt fingerprint.
Versions and CSV export
Freeze numbered versions with permanent links and export CSV with sources.

Good for

  • Verify a figure against a cited public table
  • Build a sourced dataset of companies or products from an agent
  • Classify rows in a table using your own model
  • Load a table into pandas or Google Sheets through a private CSV link
  • Suggest a correction to a public table with evidence

Pricing

Open source

Apache-2.0 source; prices for the hosted site not published

Prices change. Check skeletable.com before you buy.

Skeletable FAQ

Do I need an account to read tables?

No. Reading needs no account; proposing changes or building drafts needs a token from Settings.

Can an agent publish a table?

No. Agents can only write to drafts their person runs, and a person publishes.

Does Skeletable run a model itself?

No. AI columns are answered by your own agent.

What licence applies?

The source code is Apache-2.0; published table data is under CC BY 4.0.

How do I add it to Claude Code?

Run claude mcp add --transport http skeletable https://skeletable.com/mcp.

Is Skeletable free?

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

How do I add Skeletable to Claude Code?

Run this command in your terminal: claude mcp add --transport http skeletable https://skeletable.com/mcp

What are the best alternatives to Skeletable?

The closest alternatives to Skeletable are Codebase Memory, XERJ, Agent-Native Analytics, Claude Code Ultimate Guide and Google Ads + Meta Ads + SEO/GEO MCP.

Alternatives to Skeletable

  1. 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
  2. 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
  3. Open-source analytics app where an AI agent builds charts and dashboards from your data

    • Remote
    • Open source
    • 7.7K GitHub stars
    • Analytics and data
  4. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  5. Hosted MCP connector for managing Google Ads, Meta Ads, GA4 and Search Console from chat

    • 8 tools
    • Remote
    • Freemium
    • 8.1K GitHub stars
    • Marketing and SEO
  6. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory