ingest url data source

Create a new data source by fetching a public URL on the server side. Handles CSV, TSV, JSON, Excel, TXT, and PDF. Private networks (RFC 1918, loopback, cloud m

liveunknown~5s typical

mcpclaribi

$0.0020

per call

plus $0.001 platform fee per run

Open MCP server; Cracked bills routing only

Provider Upstream docs

$Set up https://cracked-ai-pearl.vercel.app/SKILL.md, then use Cracked to run mcp-com-claribi-mcp-server /ingest_url_data_source for me.

Input

Field	Type	Description
name*	string	Display name for the new data source.
url*	string	Public http(s) URL of the file to ingest. The server fetches it once at call time; the URL
format	string (csv \| tsv \| json \| xlsx \| xls \| txt)	Optional format hint. If omitted, the server infers it from the Content-Type header and UR
description	string	Optional human note stored on the data source.
wait_seconds	integer	Seconds to block waiting for preprocessing before returning. 0 returns immediately. default 0

Call it

curl

curl https://cracked-ai-pearl.vercel.app/v1/run \
  -H "Authorization: Bearer ck_live_..." -H "content-type: application/json" \
  -d '{"provider":"mcp-com-claribi-mcp-server","endpoint":"/ingest_url_data_source","input":{"name":"...","url":"...","wait_seconds":0}}'

cli

npx cracked-ai run -p mcp-com-claribi-mcp-server -e /ingest_url_data_source -i '{"name":"...","url":"...","wait_seconds":0}'

mcp

run_tool({ provider: "mcp-com-claribi-mcp-server", endpoint: "/ingest_url_data_source", input: {"name":"...","url":"...","wait_seconds":0} })

Try it

Runs against your signed-in workspace balance. Sign in if you have not.