Integration: Keenable

Web search and page fetch built for AI agents, keyless by default

Authors

Keenable

Table of Contents

• Overview
• Installation
• Usage
  • KeenableWebSearch
  • KeenableFetcher
• License

Overview

Keenable is a web search and page-fetch API built for AI agents. This integration provides two components:

• KeenableWebSearch: searches the web and returns results as Haystack Document objects plus their URLs (links), the same output shape as the SerperDevWebSearch / SearchApiWebSearch integrations, so it drops into existing web-search pipelines.
• KeenableFetcher: fetches one or more URLs and returns their main content as Haystack Document objects (extracted to clean markdown server-side, so you don’t need a separate LinkContentFetcher + HTMLToDocument step).

Keyless by default. With no API key, the components use Keenable’s keyless public endpoints — no signup required to try it. Set a KEENABLE_API_KEY to use the authenticated endpoints (required for mode="realtime" and for higher rate limits).

Installation

pip install keenable-haystack

Usage

KeenableWebSearch

Basic Example

from haystack_integrations.components.websearch.keenable import KeenableWebSearch

# No API key -> keyless public endpoint. Set KEENABLE_API_KEY to lift limits.
web_search = KeenableWebSearch(top_k=5)

result = web_search.run(query="latest developments in AI agents")
for doc in result["documents"]:
    print(doc.content)
print(result["links"])

You can pass an API key explicitly instead of using the environment variable:

from haystack.utils import Secret
from haystack_integrations.components.websearch.keenable import KeenableWebSearch

web_search = KeenableWebSearch(api_key=Secret.from_token("your-api-key"), top_k=5)

Parameters

• api_key: Keenable API key. Defaults to the KEENABLE_API_KEY environment variable; when absent the keyless public endpoint is used.
• top_k: Keep at most this many results (applied client-side). None keeps all.
• mode: Default search mode, "pro" (deeper) or "realtime" (low latency, key required). Overridable per run.
• site: Default single-domain restriction (e.g. "github.com"). Overridable per run.

run also accepts optional per-query filters: site, published_after, published_before, acquired_after, acquired_before, mode.

In a Pipeline

KeenableWebSearch is a drop-in for any web-search component — connect its documents output downstream:

from haystack import Pipeline
from haystack.components.builders import PromptBuilder
from haystack_integrations.components.websearch.keenable import KeenableWebSearch

pipe = Pipeline()
pipe.add_component("search", KeenableWebSearch(top_k=5))
pipe.add_component("prompt", PromptBuilder(template="Answer using:\n{{ documents }}"))
pipe.connect("search.documents", "prompt.documents")

KeenableFetcher

KeenableFetcher fetches web pages via Keenable and returns their main content as markdown. Pair it with KeenableWebSearch: discover URLs with search, then read the full pages with the fetcher.

from haystack_integrations.components.fetchers.keenable import KeenableFetcher

fetcher = KeenableFetcher()
result = fetcher.run(urls=["https://example.com/article"])
print(result["documents"][0].content)

Non-http(s) and private/internal URLs are rejected before sending, and (like LinkContentFetcher) failed URLs are skipped by default — set raise_on_failure=True to surface errors instead.

License

keenable-haystack is distributed under the terms of the MIT license.