Keymateapi SDK

Overview

Keymate.AI Web Search API: Enhances knowledge grounded responses by fetching URLs optimized for specific needs and performing authenticated internet searches.

Available Operations

upsert - Inserts record to Keymate Memory.
query - Queries the user's Keymate Memory.
browseurl - The plugin enables user to conduct web browsing by extracting the text content of a specified URL. It will generate title and content.
browse - Fetch any URLs without proxy it would probably fail on major websites but quicker than browseurl
search - Without proxies searches keyword on the internet and fetches urls and optimizes output
ultrafastsearch - This plugin provides 10 ultra fast search results from multiple sources giving a more comprehensive view.
gptsbrowse - Fetch memory.keymate.ai URLs
internetsearch - Conduct an internet search

upsert

It records the passed string in q parameter to keymate memory

Example Usage

import keymateapi

s = keymateapi.Keymateapi(
    bearer_auth="<YOUR_BEARER_TOKEN_HERE>",
)

res = s.upsert(q='I prefer Costa over Starbucks.')

if res.object is not None:
    # handle response
    pass

Parameters

Parameter	Type	Required	Description	Example
`q`	str	✔️	The context you are insertin to user's personal Keymate Memory history.	I prefer Costa over Starbucks.

Response

operations.UpsertResponse

Errors

Error Object	Status Code	Content Type
errors.SDKError	4xx-5xx	/

query

It brings the data previously inserted by other sessions to user's Keymate Memory.

Example Usage

import keymateapi

s = keymateapi.Keymateapi(
    bearer_auth="<YOUR_BEARER_TOKEN_HERE>",
)

res = s.query(q='https://keymate.ai')

if res.object is not None:
    # handle response
    pass

Parameters

Parameter	Type	Required	Description	Example
`q`	str	✔️	The context you are searching from user's personal Keymate Memory history.	https://keymate.ai

Response

operations.QueryResponse

Errors

Error Object	Status Code	Content Type
errors.SDKError	4xx-5xx	/

browseurl

This is the most powerful browsing endpoints it uses residential proxies and bypasses firewalls. Try this with Reddit, LinkedIn etc.

Example Usage

import keymateapi
from keymateapi.models import errors, operations

# Get your API token at https://my.keymate.ai/pricing
s = keymateapi.Keymateapi(
    bearer_auth="",
)

item_url = "https://www.rona.ca/en/product/danby-12-000-btu-8-000-sacc-115-v-white-portable-air-conditioner-dpa080b1wdb-6-330987187"
req = operations.BrowseurlRequest(inputwindowwords="5000", q=item_url, percentile="1", numofpages="1", paging="1")
res = s.browseurl(req)

def handle_response(response):
    content_type = response.http_res.headers.get('Content-Type', '')
    
    # Attempt to parse the response as JSON regardless of Content-Type
    try:
        response_json = response.http_res.json()
        return response_json
    except ValueError:
        # If JSON parsing fails, fall back to handling as plain text
        if 'text/plain' in content_type:
            return {'text': response.http_res.text}
        else:
            raise errors.SDKError(f'Unknown content-type received: {content_type}', response.http_res.status_code, response.http_res.text, response.http_res)

if res is not None and hasattr(res, 'object'):
    # handle response
    try:
        parsed_response = handle_response(res)
        print(res)
    except errors.SDKError as e:
        print(f"An error occurred: {e}")

print(res)

Parameters

Parameter	Type	Required	Description
`request`	operations.BrowseurlRequest	✔️	The request object to use for the request.

Response

operations.BrowseurlResponse

Errors

Error Object	Status Code	Content Type
errors.BrowseurlResponseBody	400	application/json
errors.SDKError	4xx-5xx	/

browse

Fetches URLs optimized for non heavily guarded websites

Example Usage

import keymateapi

s = keymateapi.Keymateapi(
    bearer_auth="<YOUR_BEARER_TOKEN_HERE>",
)

res = s.browse(q='http://impressive-silence.info', percentile='1', numofpages='1', paging='1')

if res.two_hundred_application_json_object is not None:
    # handle response
    pass

Parameters

Parameter	Type	Required	Description	Example
`q`	str	✔️	URL starting with https://memory.keymate.ai. Must be a valid URL.
`percentile`	str	✔️	For adjusting response scope in case of 'ResponseTooLarge' error. Starts with 1.	1
`numofpages`	str	✔️	Specifies the number of pages to return. Starts with 1 by default.	1
`paging`	Optional[str]	➖	Used for pagination. Increments for subsequent pages.	1

Response

operations.BrowseResponse

Errors

Error Object	Status Code	Content Type
errors.BrowseResponseBody	400	application/json
errors.BrowseResponseResponseBody	401	application/json
errors.SDKError	4xx-5xx	/

search

Searches web using google and fetches URLs optimized for non heavily guarded websites

Example Usage

import keymateapi

s = keymateapi.Keymateapi(
    bearer_auth="<YOUR_BEARER_TOKEN_HERE>",
)

res = s.search(q='http://fake-flume.name', percentile='1', numofpages='1')

if res.two_hundred_application_json_object is not None:
    # handle response
    pass

Parameters

Parameter	Type	Required	Description	Example
`q`	str	✔️	URL starting with https://memory.keymate.ai. Must be a valid URL.
`percentile`	str	✔️	For adjusting response scope in case of 'ResponseTooLarge' error. Starts with 1.	1
`numofpages`	str	✔️	Specifies the number of pages to return. Starts with 1 by default.	1

Response

operations.SearchResponse

Errors

Error Object	Status Code	Content Type
errors.SearchResponseBody	400	application/json
errors.SearchResponseResponseBody	401	application/json
errors.SDKError	4xx-5xx	/

ultrafastsearch

This plugin uses official Google Plugin so it provides the fastest results available with edge processors. Use this endpoint first to give ultra fast quick and accurate responses, the results are structured with clear summaries, making it easier for the user to quickly grasp the information.

Example Usage

import keymateapi

s = keymateapi.Keymateapi(
    bearer_auth="<YOUR_BEARER_TOKEN_HERE>",
)

res = s.ultrafastsearch(q='https://keymate.ai', percentile='100', numofpages='10')

if res.two_hundred_application_json_object is not None:
    # handle response
    pass

Parameters

Parameter	Type	Required	Description	Example
`q`	str	✔️	URL of the website.	https://keymate.ai
`percentile`	str	✔️	Set it as '100'	100
`numofpages`	str	✔️	Set it as '10'	10

Response

operations.UltrafastsearchResponse

Errors

Error Object	Status Code	Content Type
errors.UltrafastsearchResponseBody	400	application/json
errors.SDKError	4xx-5xx	/

gptsbrowse

Fetches URLs optimized for https://memory.keymate.ai, requiring bearer token authentication. Reflects user info and provides contextually relevant rules for actions performed.

Example Usage

import keymateapi

s = keymateapi.Keymateapi(
    bearer_auth="<YOUR_BEARER_TOKEN_HERE>",
)

res = s.gptsbrowse(q='http://puzzled-advertisement.com', percentile='1', numofpages='1', paging='1')

if res.two_hundred_application_json_object is not None:
    # handle response
    pass

Parameters

Parameter	Type	Required	Description	Example
`q`	str	✔️	URL starting with https://memory.keymate.ai. Must be a valid URL.
`percentile`	str	✔️	For adjusting response scope in case of 'ResponseTooLarge' error. Starts with 1.	1
`numofpages`	str	✔️	Specifies the number of pages to return. Starts with 1 by default.	1
`paging`	Optional[str]	➖	Used for pagination. Increments for subsequent pages.	1

Response

operations.GptsbrowseResponse

Errors

Error Object	Status Code	Content Type
errors.GptsbrowseResponseBody	400	application/json
errors.GptsbrowseResponseResponseBody	401	application/json
errors.SDKError	4xx-5xx	/

internetsearch

Performs an internet search based on provided query. Utilizes 'Authorization' and custom headers for user identification and search customization.

Example Usage

import keymateapi
import json
from keymateapi.models import errors, operations

# Get your API token at https://my.keymate.ai/pricing
s = keymateapi.Keymateapi(
    bearer_auth="<YOUR_API_KEY>", #DONT FORGET TO GET YOUR API KEY FROM https://my.keymate.ai/pricing requires paid plan, insert to this string
)
try:
    res = s.internetsearch(inputwindowwords="5000",q="python", percentile="1", numofpages="1")
    if res.two_hundred_application_json_object is not None:
        # handle response
        print(res.two_hundred_application_json_object)
    else:
        print(res)
except errors.SDKError as e:
    print("broken json")
    res = str(e).split("Status 200")[1]
    dict_data = json.loads(res)
    print("result in python dict_data start")
    print(f"{dict_data}")
    print("result in python dict_data end")

Parameters

Parameter	Type	Required	Description	Example
`inputwindowwords`	str	✔️	Set it as '8000' first if responsetoolarge occurs reduce it to 1000.	8000
`q`	str	✔️	Search query	python
`percentile`	str	✔️	Start it as '1', increase to '6' if ResponseTooLarge occurs, only reduce to '3' or '4' if user requests it.	1
`numofpages`	str	✔️	Start it as '6'. Retry the request by decreasing only this one if 'ResponseTooLarge' occurs. Should be between 1 and 10.	6

Response

operations.InternetsearchResponse

Errors

Error Object	Status Code	Content Type
errors.InternetsearchResponseBody	400	application/json
errors.InternetsearchResponseResponseBody	401	application/json
errors.SDKError	4xx-5xx	/

Files

README.md

Latest commit

History

README.md

File metadata and controls

Keymateapi SDK

Overview

Available Operations

upsert

Example Usage

Parameters

Response

Errors

query

Example Usage

Parameters

Response

Errors

browseurl

Example Usage

Parameters

Response

Errors

browse

Example Usage

Parameters

Response

Errors

search

Example Usage

Parameters

Response

Errors

ultrafastsearch

Example Usage

Parameters

Response

Errors

gptsbrowse

Example Usage

Parameters

Response

Errors

internetsearch

Example Usage

Parameters

Response

Errors