Connect an agent over MCP
Point an MCP-speaking agent — Claude, or your own — at the swarmsDB endpoint. It discovers the catalogue and calls it as native tools. This is the path a keyless OAuth connector uses, so no API key changes hands.
Platform / Access
mcp: app.swarmsdb.ai/mcp · rest: bearer + select= Swarm phase Delivery — authorized transit: the canonical dataset flies to the agent; two ways in, one surface.
Point an MCP-speaking agent — Claude, or your own — at the swarmsDB endpoint. It discovers the catalogue and calls it as native tools. This is the path a keyless OAuth connector uses, so no API key changes hands.
Prefer plain HTTP? The same capabilities answer to a REST call with a bearer token and a projection. Nothing about the response changes — you still get typed sections, not a raw upstream blob.
# claude_desktop_config.json — no API key changes hands
{ "mcpServers": {
"swarmsdb": { "url": "https://app.swarmsdb.ai/mcp" }
} } { "key_material": null,
"grant": "workspace-bound",
"revocable": true } curl https://app.swarmsdb.ai/data/openalex/works \
-H "Authorization: Bearer $SWARMSDB_KEY" \
-G --data 'search=graph neural networks' \
--data 'select=id,doi,title,cited_by_count' { "shape": "table", "rows": 25,
"fields": ["id", "doi", "title", "cited_by_count"] } # one call, many parameter sets — results by signed webhook
POST /bulk/requests
{ "endpoint": "eurostat/data",
"params": ["nrg_bal_c", "demo_gind", "prc_hicp"],
"schedule": "0 6 * * *" } { "bulk_request_id": "brq_9f2c…",
"webhook": "hmac-signed" } More in this area
Connect over MCP or REST. If your agents read more than one source, the context bill is the first thing you will see move.