Answer a real question

Go from a question to funding rounds in three calls, on whichever surface you use
View as Markdown

The quickstarts each end at a working call. This page ends at an answer.

The question: which seed-stage companies in Austin are in the dataset, and what has one of them raised? Every call below is a public read — no credential, no signup — and the three surfaces run the same three operations, so you can follow whichever column you work in.

1

Narrow the dataset to the companies you mean

Filters are the entity list’s query parameters, and they combine. stage and headquartersCity are two of more than sixty on this one operation.

curl "https://api.aventure.vc/v1/entities?headquartersCity=Austin&stage=Seed&size=5"

totalElements is the size of the whole matching set, not of your page — that number is the answer to “how many are there”. content holds the five records, each with a slug you can carry into the next step.

Location filters take the name, not the postal abbreviation: headquartersState=Texas matches, headquartersState=TX returns zero rows. Pagination and filtering covers how the filters combine.

2

Open one record

Take a slug from the previous response — healthkey-austin-tx-us below — and read the full record. The detail response carries names, locations, classifications, text summaries, and a core.id, which is the UUID the round-level operations key on.

curl "https://api.aventure.vc/v1/entities/detail?slug=healthkey-austin-tx-us"

The wire parameter is slug on all three surfaces. The CLI qualifies it as --entity-slug so that one command can also accept the --batch body field named slug; the CLI quickstart covers that distinction.

3

Read the funding rounds

core.id from step 2 is the entityId path parameter here.

curl "https://api.aventure.vc/v1/entities/019efcaf-7aff-7264-afe1-425afc1e3e11/fundraise-rounds?size=3"

Each round carries round, amountRaised, currency, and dateAnnounced:

{
"content": [
{
"id": "cb71e324-28ee-4114-b1f5-21a66a5dc78c",
"round": "Seed",
"amountRaised": 500000,
"currency": "USD",
"dateAnnounced": "2025-01-01T00:00:00.000Z"
}
],
"totalElements": 1
}

amountRaised is denominated in its own currency, so compare rounds only after converting — the values are not normalized to USD.

The shape of every other question

Those three steps are the pattern: filter a list, resolve one record, read what hangs off it. People and news work the same way, and the operations that hang off a record — news, employees, investors, similar companies — all key on the same core.id.

When you have a phrase rather than a filter set, swap step 1 for search: POST /v1/search/all runs entities, people, and news together, and aventure search --query "..." is the CLI form. Search needs a credential and a Pro subscription or trial; the three filtered calls above need neither.

To go from an API reference operation to the command that runs it, aventure docs api <operationId> prints the route, the required flags, and the matching MCP tool without calling the API.

Next

  • Pagination and filtering — the filters behind step 1, and how to page a large result set
  • Authentication — what a credential adds beyond these public reads
  • Errors — the problem detail body, and reading traceId out of a failure
  • API reference — every operation, parameter, and schema