SDKs

The official SDKs give you web search and page fetch in Python, TypeScript and Go, with results already shaped for a prompt.

What you get

search returns ranked pages with their text extracted. fetch returns one page as markdown.
to_context() renders results as numbered [n] Title (url) sources a model can cite.
TOOLS holds OpenAI-format definitions for keenable_search and keenable_fetch. A helper runs whichever tool the model calls.
In Python, AsyncKeenable has the same methods as the sync client.
Every call works with no account and no key. An API key raises your rate limit.

Install

Install the package for your language:

pip install keenable

npm install keenable

go mod init example.com/keenable-demo
go get github.com/keenableai/keenable-sdk/go@v0.1.1

The TypeScript SDK needs Node 18 or newer. The Go SDK needs Go 1.22 or newer. Neither has runtime dependencies. On macOS and many Linux distributions, system Python refuses a global pip install. Install the Python package into a virtual environment: run python3 -m venv .venv, then source .venv/bin/activate. If your Go project already has a go.mod, skip go mod init.

Run a keyless search

This program runs one search without a key and prints the top 3 results as a citable block:

from keenable import Keenable

client = Keenable()  # keyless: no API key needed to try it
results = client.search("fastest inference providers 2026", max_results=3)
print(results.to_context())

import { Keenable } from "keenable";

async function main() {
  const client = new Keenable(); // keyless: no API key needed to try it
  const results = await client.search("fastest inference providers 2026", { maxResults: 3 });
  console.log(results.toContext());
}

main();

package main

import (
	"context"
	"fmt"
	"log"

	"github.com/keenableai/keenable-sdk/go"
)

func main() {
	client, err := keenable.New(nil) // keyless: no API key needed to try it
	if err != nil {
		log.Fatal(err)
	}
	defer client.Close()

	results, err := client.Search(context.Background(), "fastest inference providers 2026",
		&keenable.SearchOptions{MaxResults: 3})
	if err != nil {
		log.Fatal(err)
	}
	fmt.Println(results.ToContext(nil))
}

curl -X POST https://api.keenable.ai/v1/search/public \
  -H "Content-Type: application/json" \
  -H "X-Keenable-Title: My App" \
  -d '{"query": "fastest inference providers 2026", "max_results": 3}'

Save it in the directory where you installed the SDK and run it:

Language	Save as	Run
Python	`main.py`	`python main.py`
TypeScript	`index.ts`	`npx tsx index.ts`
Go	`main.go`	`go run .`
curl		Paste the command into a terminal.

You see one [n] Title (url) header per result, followed by the page text:

Example output · results vary

[1] Hugging Face Inference Providers · Supported Models (https://huggingface.co/inference)
…

[2] Fastest LLM Inference Provider 2026: TTFT, tok/s, p99 (https://openbenchmarks.com/inference/fastest-inference-provider) - published 2026-09-03T00:00:00Z
Baseten recorded the lowest end-to-end p99 latency at 3.23 seconds on this run, with a median of 331 ms…

[3] Best LLM Inference Providers Compared for 2026 - SiliconFlow (https://siliconflow.com/articles/the-best-inference-provider-for-llms)
…

To raise your limit, create an API key and set KEENABLE_API_KEY. The code stays the same.

Configuration

Every SDK reads two environment variables:

Variable	Default	Purpose
`KEENABLE_API_KEY`	unset	Your API key. Unset means keyless.
`KEENABLE_API_URL`	`https://api.keenable.ai`	The API base URL.

You can also pass settings to the constructor:

Setting	Python	TypeScript	Go (`keenable.Options`)
API key	`api_key`	`apiKey`	`APIKey` (pointer)
Base URL	`base_url`	`baseUrl`	`BaseURL` (pointer)
Timeout	`timeout`, in seconds (default 30)	`timeoutMs` (default 30000)	`Timeout` (default 30s)
Client source	`client_source`	`clientSource`	`ClientSource`

The SDK sends the client source as the X-Keenable-Title header. It defaults to Keenable SDK. If you build an integration on top of an SDK, set it to your product’s name so Keenable attributes your traffic to your product.

Methods

What it does	Python	TypeScript	Go
Search the web	`search(query, **filters)`	`search(query, options)`	`Search(ctx, query, *SearchOptions)`
Fetch one page	`fetch(url)`	`fetch(url)`	`Fetch(ctx, url)`
Render a citable block	`to_context()`	`toContext()`	`ToContext(nil)`
List the results a block cites	`cited()`	`cited()`	`Cited(nil)`
Tool definitions	`TOOLS`	`TOOLS`	`Tools`
Run a tool call	`run_tool_call` (`arun_tool_call` for async)	`runToolCall`	`RunToolCall`

Search filters follow the Search parameters, in each language’s naming style: published_after in Python, publishedAfter in TypeScript, PublishedAfter in Go. SDK 0.1.x has three gaps:

In Python and TypeScript, mode accepts only pro.
No SDK sends disable_cards.
fetch takes only a URL. For live, prompt and max_chars, call the HTTP API as shown on Fetch.

Errors

Every SDK raises typed errors that share one base class. See Errors for what each status code means and how to fix it.

When	Python	TypeScript	Go
Base class	`KeenableError`	`KeenableError`	`keenable.Error` interface
HTTP 429, rate limit exceeded	`KeenableRateLimitError`	`KeenableRateLimitError`	`RateLimitError`
HTTP 401 or 403, key rejected	`KeenableAuthError`	`KeenableAuthError`	`AuthError`
Any other non-2xx response	`KeenableAPIError` (`status_code`)	`KeenableAPIError` (`statusCode`)	`APIError` (`StatusCode`)
API could not be reached	`KeenableConnectionError`	`KeenableConnectionError`	`ConnectionError`
Bad arguments, no request sent	`KeenableInvalidRequestError`	`KeenableInvalidRequestError`	`InvalidRequestError`

In 0.1.x, an HTTP 402 (out of credits) raises the generic API error. To catch it, check status_code == 402 in Python, statusCode === 402 in TypeScript, or StatusCode == 402 on an APIError in Go.

Source, packages and changelogs

Language	README	Package	Changelog
Python	python/README.md	PyPI	python/CHANGELOG.md
TypeScript	typescript/README.md	npm	typescript/CHANGELOG.md
Go	go/README.md	pkg.go.dev	go/CHANGELOG.md

Get started

Products

Connect

Resources

What you get

Install

Run a keyless search

Configuration

Methods

Errors

Source, packages and changelogs

​What you get

​Install

​Run a keyless search

​Configuration

​Methods

​Errors

​Source, packages and changelogs

What you get

Install

Run a keyless search

Configuration

Methods

Errors

Source, packages and changelogs