National Capability Benchmark
Read-only country capability scores with confidence ratings for 135 countries
Answered our latest tool-list check 3 hours ago with 6 tools. It will be marked verified after 3 successful checks in 7 days.
- Platforms
- API, Web
- Runs
- Remotely
- Version
- 1.0.0
- Listing updated
- today
Similar MCP servers
About National Capability Benchmark
National Capability Benchmark is an MCP server for analytics and data. It runs as a hosted remote server, so there is nothing to install locally.
The NCB project scores 135 countries across nine capability dimensions, such as anticipating change, coordinating around it, and building what a country decides to build. Each score sits on a 0-to-100 scale set by the benchmark countries as a whole, and each comes with a confidence value that drops when the underlying evidence is thin. The site describes itself as a research prototype from Envisioning and says it is designed to test whether these capabilities are more than a country's income.
The project publishes its dataset (version 10.2.0), per-capability country rankings, comparisons with income peers, and the indicators behind each score. Its MCP server exposes these scores over streamable HTTP at a remote endpoint and is read-only. Governments, development agencies, researchers and journalists are the audiences the site names.
Because scores are positions within a frame rather than percentages, the site advises against reading a low number as a literal share of a capability. Averaging the nine dimensions can also hide large gaps between them, which is why the project shows each one separately.
Install
Run in a terminal
claude mcp add --transport http ncb https://ncb.envisioning.com/mcpAdd to Settings → Connectors → Add custom connector
https://ncb.envisioning.com/mcpAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"ncb": {
"url": "https://ncb.envisioning.com/mcp"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"ncb": {
"type": "http",
"url": "https://ncb.envisioning.com/mcp"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"ncb": {
"serverUrl": "https://ncb.envisioning.com/mcp"
}
}
}Add to Grok Bot desktop app → Connect apps → Add a custom connector
https://ncb.envisioning.com/mcpEnter a Name (for example "National Capability Benchmark"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.
Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"ncb": {
"type": "http",
"url": "https://ncb.envisioning.com/mcp"
}
}
}Using several servers? Build one config file with the MCP config generator.
Tools this server provides (6)
Read directly from the server.
All 6 tools say they're read-only.
From the server's own tool annotations; not verified by us.
- list_dimensionsSays read-only
- The nine capability dimensions, the question each one answers, and how many indicators each has: scored, declared gaps, rejected, and conditions published beside it. Also the dataset version and the quoting contract. Start here.
- get_countrySays read-only
- One country's nine dimensions, each with score, confidence and its band, indicator count, coverage flag, trend, panel estimate where one exists, and links to the known limits that apply. Takes an ISO3 code or an English country name.
- compare_countriesSays read-only
- Two to 4 countries side by side on the nine dimensions, in the order you give them. The result is not sorted by score: the benchmark compares shapes and does not rank (D1).
- search_indicatorsSays read-only
- Search the registry of indicators by words in the name or definition, and filter by dimension or status: scored, gap (no adequate dataset yet), rejected (a dataset exists and the project disqualified it) or condition (published beside a dimension, never scored).
- get_decisionSays read-only
- Read one entry of the decision record by number, for example D1 or D118: the methodological choice, why it was made, what it costs and what evidence would overturn it. Use it to answer "why does the benchmark do X?".
- get_limitSays read-only
- Read one known limit by id (A1, A2, ...): a place where the model produces a number that is wrong about the world. get_country lists the ids that apply to each score. Without an id, lists every limit with the dimensions and countries it covers.
What it does
- Nine capability scores
- Each of 135 countries receives a 0-100 position on nine separate dimensions.
- Confidence ratings
- Every score carries a confidence value, and weakly evidenced scores are shown with broken rings.
- Income comparison
- Countries can be set beside their nearest income peers to see how capability differs from income.
- Read-only remote endpoint
- The MCP server is served over streamable HTTP and only reads data.
- Citable dataset versions
- Results are tied to a numbered dataset release with a stated citation format.
Good for
- Compare a country's capability profile with its income peers
- Find which capabilities a government has below 50 on usable evidence
- Pull country capability scores into an AI client for research or reporting
- Check whether capability patterns hold up once income is accounted for
- Cite a specific capability figure with its dataset version
National Capability Benchmark FAQ
Is the MCP server read-only?
Yes. The registry entry describes it as read-only.
How many countries and dimensions does it cover?
It covers 135 countries and nine capability dimensions.
What does a score of 10 mean?
The site says a score is a position inside the benchmark frame, not a percentage of a capability, so 10 sits near the floor of that frame.
How do I add National Capability Benchmark to Claude Code?
Run this command in your terminal: claude mcp add --transport http ncb https://ncb.envisioning.com/mcp
What are the best alternatives to National Capability Benchmark?
The closest alternatives to National Capability Benchmark are Statistics of the World, Zovo Country Data, (iS)NEPI - National Energy Performance Index, Worldbank MCP and MapleStats MCP.
Where this comes from
- Description: written from the project's website and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 19 November 2026.
- Install steps: from the official MCP Registry entry.