Skip to main content
Glama
Sugra API MCP

Sugra API MCP

Official

Sugra MCP is built and run by Sugra Systems, Inc. It brings Sugra Data into AI assistants and agents over the Model Context Protocol: 1,600+ endpoints from 200 primary sources across 36 data domains - central banks, national statistics offices, regulators, intergovernmental bodies and more. The agent searches the catalog, reads an endpoint's description and calls it, and every answer carries its source and date.

Published in Anthropic's Connectors Directory, the OpenAI Plugins Directory and the GitHub MCP Registry. Works with Claude, Claude Code, ChatGPT, Codex, Grok, Cursor, Antigravity, GitHub Copilot, VS Code, Zed, Cline, Continue and any MCP client.

Watch: How Sugra MCP works · Seven directions, one key

Seven directions

Direction

What an agent can ask for

Learn more

Sugra Finance

Quotes, price history, company financials, SEC filings, funds, options, futures and Treasuries

Docs · Video · Article

Sugra Macro

Inflation, GDP, interest rates, jobs, trade and public finance of whole countries

Docs · Video · Article

Sugra News

Headlines, feeds and the world's media attention

Docs · Video · Article

Sugra Earth

Weather, climate, air quality, energy, natural hazards and vessel traffic

Docs · Video · Article

Sugra NetAtlas

IP and ASN lookups, routing history, geolocation and internet outages

Docs · Video · Article

Sugra Entity

Legal entity identifiers, company registries, sanctions and PEP screening

Docs · Video · Article

Sugra Research

Papers, preprints, citations and authors

Docs · Article

Related MCP server: @prism/mcp-server

Install

The server is hosted, so there is nothing to run. Add Sugra from your agent's directory or install the sugra-api-skills plugin, then connect as shown below. The plugin also carries the Sugra API Skills, which give the agent the procedure - find an operation, read the response, cite the source.

Claude (claude.ai, Desktop, mobile and Cowork): Add to Claude opens Sugra in Anthropic's Connectors Directory.

Claude Code, the plugin with the skills and the server:

/plugin marketplace add Sugra-Systems/sugra-api-plugins
/plugin install sugra-api-skills@sugra-api-plugins

Add to ChatGPT opens Sugra in the OpenAI Plugins Directory, which serves both ChatGPT and Codex. In ChatGPT the skills install together with the server.

Codex, the skills plugin:

codex plugin marketplace add Sugra-Systems/sugra-api-plugins
codex plugin add sugra-api-skills@sugra-api-plugins

GitHub MCP Registry opens Sugra in GitHub's registry. Its Install MCP server button adds the server to VS Code, VS Code Insiders or the GitHub Copilot app.

GitHub Copilot CLI, the plugin with the skills and the server:

copilot plugin marketplace add Sugra-Systems/sugra-api-plugins
copilot plugin install sugra-api-skills@sugra-api-plugins

VS Code: Install in VS Code adds the server.

Grok, the plugin with the skills and the server:

grok plugin install Sugra-Systems/sugra-api-plugins#xai

Cursor: Install in Cursor adds the server. The plugin with the skills and the server:

git clone https://github.com/Sugra-Systems/sugra-api-plugins
mkdir -p ~/.cursor/plugins/local
cp -R sugra-api-plugins/cursor ~/.cursor/plugins/local/sugra-api-skills

The skills plugin:

git clone https://github.com/Sugra-Systems/sugra-api-plugins
agy plugin install ./sugra-api-plugins/antigravity

Antigravity does not start the server from the plugin: add it as shown under Connect.

The skills, for any agent that reads Agent Skills:

npx skills add https://mcp.sugra.ai

The server: add it as shown under Connect.

Run it locally

The PyPI package runs the server on your machine over stdio with an API key. Python 3.13 or newer:

pip install sugra-api-mcp
export SUGRA_API_KEY=sugra_...        # free key: https://app.sugra.ai/register
sugra-api-mcp call quotes_symbol_price --params '{"symbol":"AAPL"}'

Connect

Every client connects to one address. Sign in with your Sugra account, or send an API key as Authorization: Bearer <key>:

https://mcp.sugra.ai/mcp

Claude: connect Sugra in the directory and sign in with your Sugra account. Added Sugra earlier as a custom connector? Remove that one before you connect from the directory, or you will have two.

Claude Code signed in with a claude.ai account picks the connector up on its own. With the plugin, sign in to its server from /mcp.

In ChatGPT or Codex, connect Sugra and sign in with your Sugra account.

GitHub Copilot CLI, by hand:

copilot mcp add --transport http sugra-api https://mcp.sugra.ai/mcp

VS Code, by hand, .vscode/mcp.json:

{ "servers": { "sugra": { "type": "http", "url": "https://mcp.sugra.ai/mcp" } } }

Both open the Sugra sign-in page in your browser the first time they connect.

Grok connects through the plugin.

Cursor, by hand, ~/.cursor/mcp.json:

{ "mcpServers": { "sugra": { "url": "https://mcp.sugra.ai/mcp" } } }

To sign in, open Cursor Settings, then Tools & MCPs, and select Connect next to sugra.

agy mcp add sugra-api https://mcp.sugra.ai/mcp

The CLI, the app and the IDE share the server. Antigravity opens the Sugra sign-in page the first time the agent uses it.

Add the address above as a remote MCP server in any client that speaks Streamable HTTP.

{
  "mcpServers": {
    "sugra": {
      "command": "sugra-api-mcp",
      "env": { "SUGRA_API_KEY": "sugra_..." }
    }
  }
}

Claude Desktop reads it from claude_desktop_config.json, Cursor from ~/.cursor/mcp.json. Other stdio clients take the same command and environment variable in their own MCP settings.

Tools

Tool

What it does

fetch_data

Finds the best endpoint for a plain-language request and calls it, in one step

search_endpoints

Searches the catalog of 1,600+ endpoints

describe_endpoint

Shows an endpoint's parameters, required inputs and cost hints

call_endpoint

Calls an endpoint by its operation id, with optional limit and fields

list_toolsets

Lists the catalog groups with their endpoint counts

list_sources

Lists the sources behind the catalog

sugra_entity_screen

Screens a name against sanctions lists and watchlists

sugra_entity_lookup

Looks up a company by LEI or VAT number, with screening

resolve_entity

Hosted only: turns a ticker, company, indicator, coin or currency pair into one entity

get_snapshot

Hosted only: one current view of an entity, with freshness and provenance

get_timeseries

Hosted only: a bounded series of an entity's price or macro metric

The server also ships six workflow prompts (market snapshot, macro briefing, sanctions screening, sector comparison, earth conditions, source overview) and serves five of the Sugra API Skills as MCP resources under sugra://skills/.

Documentation

Blog

Development

git clone https://github.com/Sugra-Systems/sugra-api-mcp
cd sugra-api-mcp
pip install -e ".[dev,http]"
pytest

The test matrix runs on Python 3.13, 3.14 and 3.15.

Help

License

MIT &copy; 2026 Sugra Systems, Inc.

Available Tools

8 tools
call_endpointA
Read-onlyIdempotent
Inspect

Call a Sugra API endpoint by operation_id from the bundled catalog.

Plan calls with describe_endpoint's agent_hints: duration_class "fast" usually responds in under ~2s, "slow" usually 1-5s and occasionally 15s+ on a cold upstream, "heavy" can exceed the gateway timeout - keep parallel calls within max_concurrency and prefer small batches. Bulk endpoints bill 1 request credit per body item. Failures return structured errors {error, reason, status_code, elapsed_ms, retry_hint}; after "upstream_timeout" a single retry often succeeds because the aborted attempt warms upstream caches.

Inputs: operation_id (required), body, limit, fields, +2 more. Returns: object

describe_endpointA
Read-onlyIdempotent
Inspect

Describe one Sugra API endpoint by operation_id.

Includes agent_hints (duration_class fast/slow/heavy, max_concurrency, bulk billing) so you can budget timeouts and parallelism before calling. POST endpoints with a JSON body also carry request_body_schema (the resolved JSON schema) - construct the body argument from it instead of guessing key names. Call this after search_endpoints and before call_endpoint when you need the exact parameter names and examples.

Inputs: operation_id (required). Returns: object

fetch_dataA
Read-onlyIdempotent
Inspect

One-step fetch: find the best Sugra endpoint for the query and call it.

Combines search_endpoints + call_endpoint into a single round trip. Use this when you want data without manually picking an operation_id. The full search_endpoints + describe_endpoint + call_endpoint dance is still available when you need explicit control, but for most natural-language queries this tool is enough.

Behavior:

  1. Search the bundled catalog for the query. Top match wins.

  2. If the matched endpoint has required parameters and they are all provided in params, call it and return the response.

  3. If required parameters are missing, return the candidate endpoints and the missing-params list so the LLM can retry with the correct params dict on the next call.

  4. If the query names a country and the match takes a country or countries param that params leaves unset, return needs_params for it with query_countries (ISO2) instead of running the match without that filter.

Examples:

  • fetch_data("US CPI inflation", params={"series_id": "CPIAUCSL"}) runs fred_series_series_id (/api/v1/fred/series/CPIAUCSL) and returns observations.

  • fetch_data("Bitcoin price") runs onchain_bitcoin_price, which takes no params.

  • fetch_data("Latest financial news") runs news_latest, which has no required params. Only the top match runs, and a param it does not declare returns error unknown_parameters. For one coin's price, name the operation: call_endpoint("crypto_coin_id_price", params={"coin_id": "bitcoin"}).

Inputs: query (required), body, limit, fields, +2 more. Returns: object

list_sourcesA
Read-onlyIdempotent
Inspect

List source families in the bundled catalog with endpoint counts.

Use the family names as the source filter on search_endpoints. This does not call the Sugra API.

Inputs: no named parameters. Returns: object

list_toolsetsA
Read-onlyIdempotent
Inspect

List catalog groups with endpoint counts and short descriptions.

Use the group names as the toolset filter on search_endpoints. This does not call the Sugra API; it reads the bundled catalog.

Inputs: no named parameters. Returns: object

search_endpointsA
Read-onlyIdempotent
Inspect

Search the bundled Sugra endpoint catalog by natural-language query.

Use this to pick an operation_id. It does not fetch data. Typical loop:

  1. search_endpoints(query) -> ranked hits with required_parameters

  2. describe_endpoint(operation_id) -> params, request_body_schema, agent_hints

  3. call_endpoint(operation_id, params=..., body=...) or fetch_data(query, params=...)

Filter with toolset or source only after list_toolsets / list_sources; a misspelled filter is an error, not a silent empty result.

Examples:

  • search_endpoints("US CPI inflation")

  • search_endpoints("AAPL price", toolset="markets")

  • search_endpoints("container ship AIS", toolset="network")

Inputs: query (required), limit, source, toolset. Returns: object

sugra_entity_lookupA
Read-onlyIdempotent
Inspect

Resolve an entity by identifier and return its composed KYB envelope.

anchor is lei (Legal Entity Identifier, resolved via the GLEIF registry) or vat (EU VAT number, validated via the EU VIES service). The result weaves identity, a sanctions screening signal, and - on request - ownership and adverse-media slices.

The screening verdict is a SCREENING SIGNAL, not a compliance determination, and any PEP / adverse-media content is supplementary and non-comprehensive. The disclaimer field carries this and is always present.

Output is COMPACT by default to protect the agent context budget: {entity:{name, anchor, value, status, country}, screening:{status, top_matches:[...3], hit_count}, ids:{...}, disclaimer}. Pass include to opt INTO fuller per-slice detail, e.g. include=["ownership","adverse_media"] adds those slices in full form.

On a bad anchor or an API error this returns a clean {error, detail} dict rather than raising, so the agent can branch on result.get("error").

Args: anchor: Identifier type, one of lei or vat. value: The identifier value (the 20-char LEI code or the VAT number). include: Optional list of fuller slices to add, e.g. ["ownership", "adverse_media"]. Omit for the compact default.

Inputs: value (required), anchor (required), include. Returns: object

sugra_entity_screenA
Read-onlyIdempotent
Inspect

Screen a person or organization name against the Sugra sanctions corpus.

Returns a SCREENING SIGNAL, not a compliance determination. Sugra is a technology provider, not a sanctions authority or consumer reporting agency. PEP and adverse-media coverage is supplementary and non-comprehensive - a clear result is not proof of absence, and a hit is a candidate match to review, not a finding.

Output is COMPACT to protect the agent context budget: {status, matches:[{name, score, list, type}], disclaimer}. The verdict status is one of clear, review, or hit. The heavy raw fields (match rationale, source ids, publish dates) are dropped; use the Sugra API directly when the full screening envelope is needed.

Args: name: The person or organization name to screen (required). country: Optional ISO 3166-1 alpha-2 country to narrow the match. dob: Optional date of birth (YYYY-MM-DD) for a person. nationality: Optional nationality to narrow the match.

Inputs: name (required), dob, country, nationality. Returns: object

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.3/5.0

    Scored across 8 tools

    Disambiguation4/5

    Each tool has a reasonably distinct role: catalog discovery (list_toolsets, list_sources, search_endpoints, describe_endpoint), execution (call_endpoint, fetch_data), and entity screening (sugra_entity_lookup, sugra_entity_screen). The main overlap is fetch_data versus the search+describe+call sequence, and list_toolsets versus list_sources, but the descriptions explicitly explain when to prefer each, so misselection risk is low.

    Naming Consistency4/5

    All names are snake_case with a verb_noun shape (search_endpoints, call_endpoint, fetch_data, list_sources, describe_endpoint), which is consistent and readable. The two entity tools break the pattern by prefixing a product name (sugra_entity_lookup, sugra_entity_screen), a minor deviation that still conveys grouping.

    Tool Count5/5

    Eight tools is well within the ideal range and each one earns its place: discovery, description, direct calling, a one-step convenience path, and two domain-specific entity operations. Nothing appears redundant or bolted on.

    Completeness4/5

    The surface covers the full loop of finding, inspecting, and invoking catalog endpoints, plus the entity KYB/screening use case, with structured errors and parameter feedback. Minor gaps exist — no tool to enumerate all endpoints within a toolset/source without a query, and batch invocation is only implicit via call_endpoint's bulk bodies — but agents can work around these.

    Maintenance

    ActivityActive
    ResponsivenessLimited data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      B
      maintenance
      US/HK markets — 110 tools: real-time quotes, options, orders, fundamentals, alerts, DCA & portfolio
      193
      16
      Apache 2.0
    • A
      license
      Not graded
      quality
      D
      maintenance
      Enables AI agents to access real-time financial data including crypto, equities, on-chain, prediction markets, and macro via a single API key.
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      Provides live data API for AI agents with 51 REST endpoints across finance, travel, weather, news, and more, using USDC micropayments without API keys.
      42 npm
      MIT
    • A
      license
      A
      quality
      C
      maintenance
      Provides curated US economic data from Treasury, FRED, BLS, BEA, and other sources through an MCP interface. Enables querying economic series, fetching data with provenance tracking, and accessing cached artifacts.
      9
      MIT