Python Agent Tools for Graphlit Platform
Overview
The Graphlit Agent Tools for Python enables easy interaction with agent frameworks such as CrewAI, allowing developers to easily integrate the Graphlit service with agentic workflows. This document outlines the setup process and provides a basic example of using the tools.
Prerequisites
Before you begin, ensure you have the following:
- Python 3.x installed on your system.
- An active account on the Graphlit Platform with access to the API settings dashboard.
Installation
To install the Graphlit Agent Tools with CrewAI, use pip:
pip install graphlit-tools[crewai]
Using the Graphlit agent tools
We have example Google Colab notebooks using CrewAI, which provide an example for analyzing the web marketing strategy of a company, and for structured data extraction of products from scraped web pages.
Once you have configured the Graphlit client, as shown below, you will pass the client to the tool constructor.
For use in CrewAI, you will need to convert the tool to the CrewAI tool schema with the CrewAIConverter.from_tool()
function. We will provide support for additional agent frameworks, such as LangGraph and AutoGen in future.
from graphlit_tools import WebSearchTool, CrewAIConverter
web_search_tool = CrewAIConverter.from_tool(WebSearchTool(graphlit))
web_search_agent = Agent(
role="Web Researcher",
goal="Find the {company} website.",
backstory="",
verbose=True,
allow_delegation=False,
tools=[web_search_tool],
)
Configuration
The Graphlit Client supports environment variables to be set for authentication and configuration:
GRAPHLIT_ENVIRONMENT_ID
: Your environment ID.GRAPHLIT_ORGANIZATION_ID
: Your organization ID.GRAPHLIT_JWT_SECRET
: Your JWT secret for signing the JWT token.
Alternately, you can pass these values with the constructor of the Graphlit client.
You can find these values in the API settings dashboard on the Graphlit Platform.
For example, to use Graphlit in a Google Colab notebook, you need to assign these properties as Colab secrets: GRAPHLIT_ORGANIZATION_ID, GRAPHLIT_ENVIRONMENT_ID and GRAPHLIT_JWT_SECRET.
import os
from google.colab import userdata
from graphlit import Graphlit
os.environ['GRAPHLIT_ORGANIZATION_ID'] = userdata.get('GRAPHLIT_ORGANIZATION_ID')
os.environ['GRAPHLIT_ENVIRONMENT_ID'] = userdata.get('GRAPHLIT_ENVIRONMENT_ID')
os.environ['GRAPHLIT_JWT_SECRET'] = userdata.get('GRAPHLIT_JWT_SECRET')
graphlit = Graphlit()
Setting Environment Variables
To set these environment variables on your system, use the following commands, replacing your_value
with the actual values from your account.
For Unix/Linux/macOS:
export GRAPHLIT_ENVIRONMENT_ID=your_environment_id_value
export GRAPHLIT_ORGANIZATION_ID=your_organization_id_value
export GRAPHLIT_JWT_SECRET=your_secret_key_value
For Windows Command Prompt (CMD):
set GRAPHLIT_ENVIRONMENT_ID=your_environment_id_value
set GRAPHLIT_ORGANIZATION_ID=your_organization_id_value
set GRAPHLIT_JWT_SECRET=your_secret_key_value
For Windows PowerShell:
$env:GRAPHLIT_ENVIRONMENT_ID="your_environment_id_value"
$env:GRAPHLIT_ORGANIZATION_ID="your_organization_id_value"
$env:GRAPHLIT_JWT_SECRET="your_secret_key_value"
Tools
Content Ingestion
URLIngestTool: Graphlit URL ingest tool
Description
Ingests content from URL.
Returns extracted Markdown text and metadata from content.
Can ingest individual Word documents, PDFs, audio recordings, videos, images, or any other unstructured data.
Parameters
Name | Type | Description |
---|
url | str | URL of cloud-hosted file to be ingested into knowledge base |
LocalIngestTool: Graphlit local file ingest tool
Description
Ingests content from local file.
Returns extracted Markdown text and metadata from content.
Can ingest individual Word documents, PDFs, audio recordings, videos, images, or any other unstructured data.
Parameters
Name | Type | Description |
---|
file_path | str | Path of local file to be ingested into knowledge base |
WebScrapeTool: Graphlit web scrape tool
Description
Scrapes web page into knowledge base.
Returns Markdown text and metadata extracted from web page.
Parameters
Name | Type | Description |
---|
url | str | URL of web page to be scraped and ingested into knowledge base |
WebCrawlTool: Graphlit web crawl tool
Description
Crawls web pages from web site into knowledge base.
Returns Markdown text and metadata extracted from web pages.
Parameters
Name | Type | Description |
---|
url | str | URL of web site to be crawled and ingested into knowledge base |
search | Optional[str] | Text to search for within ingested web pages |
read_limit | Optional[int] | Maximum number of web pages from web site to be crawled |
WebSearchTool: Graphlit web search tool
Description
Accepts search query text as string.
Performs web search based on search query.
Returns Markdown text and metadata extracted from web pages.
Parameters
Name | Type | Description |
---|
search | str | Text to search for within web pages across the Internet |
search_limit | Optional[int] | Maximum number of web pages to be returned from web search |
WebMapTool: Graphlit web map tool
Description
Accepts web page URL as string.
Enumerates the web pages at or beneath the provided URL using web sitemap.
Returns list of mapped URIs from web site.
Parameters
Name | Type | Description |
---|
url | str | URL of the web page to be mapped |
RedditIngestTool: Graphlit Reddit ingest tool
Description
Ingests posts from Reddit subreddit into knowledge base.
Returns extracted Markdown text and metadata from Reddit posts.
Parameters
Name | Type | Description |
---|
subreddit_name | str | Reddit subreddit name to be read and ingested into knowledge base |
search | Optional[str] | Text to search for within ingested posts |
read_limit | Optional[int] | Maximum number of posts from Reddit subreddit to be read, defaults to 10 |
NotionIngestTool: Graphlit Notion ingest tool
Description
Ingests pages from Notion database into knowledge base.
Returns extracted Markdown text and metadata from Notion pages.
Requires NOTION_API_KEY to be assigned as environment variable.
Parameters
Name | Type | Description |
---|
search | Optional[str] | Text to search for within ingested pages |
read_limit | Optional[int] | Maximum number of pages from Notion database to be read, defaults to 10 |
Description
Ingests posts from RSS feed into knowledge base.
For podcast RSS feeds, audio will be transcribed and ingested into knowledge base.
Returns extracted or transcribed Markdown text and metadata from RSS posts.
Parameters
Name | Type | Description |
---|
url | str | RSS URL to be read and ingested into knowledge base |
search | Optional[str] | Text to search for within ingested posts and/or transcripts |
read_limit | Optional[int] | Maximum number of posts from RSS feed to be read, defaults to 10 |
MicrosoftEmailIngestTool: Graphlit Microsoft Email ingest tool
Description
Ingests emails from Microsoft Email account into knowledge base.
Returns extracted Markdown text and metadata from emails.
Requires MICROSOFT_EMAIL_CLIENT_ID, MICROSOFT_EMAIL_CLIENT_SECRET and MICROSOFT_EMAIL_REFRESH_TOKEN to be assigned as environment variables.
Parameters
Name | Type | Description |
---|
search | Optional[str] | Text to search for within ingested email |
read_limit | Optional[int] | Maximum number of emails from Microsoft Email account to be read, defaults to 10 |
GoogleEmailIngestTool: Graphlit Google Email ingest tool
Description
Ingests emails from Google Email account into knowledge base.
Returns extracted Markdown text and metadata from emails.
Requires GOOGLE_EMAIL_CLIENT_ID, GOOGLE_EMAIL_CLIENT_SECRET and GOOGLE_EMAIL_REFRESH_TOKEN to be assigned as environment variables.
Parameters
Name | Type | Description |
---|
search | Optional[str] | Text to search for within ingested email |
read_limit | Optional[int] | Maximum number of emails from Google Email account to be read, defaults to 10 |
GitHubIssueIngestTool: Graphlit GitHub Issue ingest tool
Description
Ingests issues from GitHub repository into knowledge base.
Accepts GitHub repository owner and repository name.
For example, for GitHub repository (https://github.com/openai/tiktoken), 'openai' is the repository owner, and 'tiktoken' is the repository name.
Returns extracted Markdown text and metadata from issues.
Requires GITHUB_PERSONAL_ACCESS_TOKEN to be assigned as environment variable.
Parameters
Name | Type | Description |
---|
repository_name | str | GitHub repository name |
repository_owner | str | GitHub repository owner |
search | Optional[str] | Text to search for within ingested issues |
read_limit | Optional[int] | Maximum number of issues from GitHub repository to be read, defaults to 10 |
JiraIssueIngestTool: Graphlit Jira ingest tool
Description
Ingests issues from Atlassian Jira into knowledge base.
Accepts Atlassian Jira server URL and project name.
Returns extracted Markdown text and metadata from issues.
Requires JIRA_TOKEN and JIRA_EMAIL to be assigned as environment variables.
Parameters
Name | Type | Description |
---|
url | str | Atlassian Jira server URL |
project | str | Atlassian Jira project name |
search | Optional[str] | Text to search for within ingested issues |
read_limit | Optional[int] | Maximum number of issues from Jira project to be read, defaults to 10 |
LinearIssueIngestTool: Graphlit Linear ingest tool
Description
Ingests issues from Linear project into knowledge base.
Accepts Linear project name.
Returns extracted Markdown text and metadata from issues.
Requires LINEAR_API_KEY to be assigned as environment variable.
Parameters
Name | Type | Description |
---|
project | str | Linear project name |
search | Optional[str] | Text to search for within ingested issues |
read_limit | Optional[int] | Maximum number of issues from Linear project to be read, defaults to 10 |
MicrosoftTeamsIngestTool: Graphlit Microsoft Teams ingest tool
Description
Ingests messages from Microsoft Teams channel into knowledge base.
Returns extracted Markdown text and metadata from messages.
Requires MICROSOFT_TEAMS_CLIENT_ID, MICROSOFT_TEAMS_CLIENT_SECRET and MICROSOFT_TEAMS_REFRESH_TOKEN to be assigned as environment variables.
Parameters
Name | Type | Description |
---|
team_name | str | Microsoft Teams team name |
channel_name | str | Microsoft Teams channel name |
search | Optional[str] | Text to search for within ingested messages |
read_limit | Optional[int] | Maximum number of messages from Microsoft Teams channel to be read, defaults to 10 |
DiscordIngestTool: Graphlit Discord ingest tool
Description
Ingests messages from Discord channel into knowledge base.
Accepts Discord channel name.
Returns extracted Markdown text and metadata from messages.
Requires DISCORD_BOT_TOKEN to be assigned as environment variable.
Parameters
Name | Type | Description |
---|
channel_name | str | Discord channel name |
search | Optional[str] | Text to search for within ingested messages |
read_limit | Optional[int] | Maximum number of messages from Discord channel to be read, defaults to 10 |
SlackIngestTool: Graphlit Slack ingest tool
Description
Ingests messages from Slack channel into knowledge base.
Accepts Slack channel name.
Returns extracted Markdown text and metadata from messages.
Requires SLACK_BOT_TOKEN to be assigned as environment variable.
Parameters
Name | Type | Description |
---|
channel_name | str | Slack channel name |
search | Optional[str] | Text to search for within ingested messages |
read_limit | Optional[int] | Maximum number of messages from Slack channel to be read, defaults to 10 |
RAG
PromptTool: Graphlit RAG prompt tool
Description
Accepts user prompt as string.
Prompts LLM with relevant content and returns completion from RAG pipeline. Returns Markdown text from LLM completion.
Uses vector embeddings and similarity search to retrieve relevant content from knowledge base.
Can search through web pages, PDFs, audio transcripts, and other unstructured data.
Parameters
Name | Type | Description |
---|
prompt | str | Text prompt which is provided to LLM for completion, via RAG pipeline |
Data Retrieval
ContentRetrievalTool: Graphlit content retrieval tool
Description
Accepts search text as string.
Optionally accepts a list of content types (i.e. FILE, PAGE, EMAIL, ISSUE, MESSAGE) for filtering the result set.
Retrieves contents based on similarity search from knowledge base.
Returns extracted Markdown text and metadata from contents relevant to the search text.
Can search through web pages, PDFs, audio transcripts, Slack messages, emails, or any unstructured data ingested into the knowledge base.
Parameters
Name | Type | Description |
---|
text | str | Text to search for within the knowledge base |
types | Optional[List[ContentTypes]] | List of content types (i.e. FILE, PAGE, EMAIL, ISSUE, MESSAGE) to be returned from knowledge base |
limit | Optional[int] | Number of contents to return from search query |
PersonRetrievalTool: Graphlit person retrieval tool
Description
Accepts search text as string.
Retrieves persons based on similarity search from knowledge base.
Returns metadata from persons relevant to the search text.
Parameters
Name | Type | Description |
---|
search | str | Text to search for within the knowledge base |
limit | Optional[int] | Number of persons to return from search query |
OrganizationRetrievalTool: Graphlit organization retrieval tool
Description
Accepts search text as string.
Retrieves organizations based on similarity search from knowledge base.
Returns metadata from organizations relevant to the search text.
Parameters
Name | Type | Description |
---|
search | str | Text to search for within the knowledge base |
limit | Optional[int] | Number of organizations to return from search query |
Image Description
DescribeImageTool: Graphlit image description tool
Description
Accepts image URL as string.
Prompts vision LLM and returns completion. Returns Markdown text from LLM completion.
Parameters
Name | Type | Description |
---|
url | str | URL for image to be described with vision LLM |
prompt | str | Text prompt which is provided to vision LLM for completion |
DescribeWebPageTool: Graphlit screenshot web page tool
Description
Screenshots web page from URL and describes web page with vision LLM.
Returns Markdown description of screenshot and extracted Markdown text from image.
Parameters
Name | Type | Description |
---|
url | str | URL of web page to screenshot and ingest into knowledge base |
prompt | Optional[str] | Text prompt which is provided to vision LLM for screenshot description |
Content Generation
GenerateSummaryTool: Graphlit summary generation tool
Description
Accepts text as string.
Optionally accepts text prompt to be provided to LLM for text summarization.
Returns summary as text.
Parameters
Name | Type | Description |
---|
text | str | Text to be summarized |
prompt | Optional[str] | Text prompt which is provided to LLM for text summarization |
GenerateBulletsTool: Graphlit bullet points generation tool
Description
Accepts text as string.
Optionally accepts the count of bullet points to be generated.
Returns bullet points as text.
Parameters
Name | Type | Description |
---|
text | str | Text to be summarized into bullet points |
count | Optional[int] | Number of bullet points to be generated |
GenerateHeadlinesTool: Graphlit headlines generation tool
Description
Accepts text as string.
Optionally accepts the count of headlines to be generated.
Returns headlines as text.
Parameters
Name | Type | Description |
---|
text | str | Text to be summarized into headlines |
count | Optional[int] | Number of headlines to be generated |
GenerateSocialMediaPostsTool: : Graphlit social media posts generation tool
Description
Accepts text as string.
Optionally accepts the count of social media posts to be generated.
Returns social media posts as text.
Parameters
Name | Type | Description |
---|
text | str | Text to be summarized into social media posts |
count | Optional[int] | Number of social media posts to be generated |
GenerateQuestionsTool: Graphlit followup questions generation tool
Description
Accepts text as string.
Optionally accepts the count of followup questions to be generated.
Returns followup questions as text.
Parameters
Name | Type | Description |
---|
text | str | Text to be summarized into followup questions |
count | Optional[int] | Number of followup questions to be generated |
GenerateKeywordsTool: Graphlit keywords generation tool
Description
Accepts text as string.
Optionally accepts the count of keywords to be generated.
Returns keywords as text.
Parameters
Name | Type | Description |
---|
text | str | Text to be summarized into keywords |
count | Optional[int] | Number of keywords to be generated |
GenerateChaptersTool: Graphlit transcript chapters generation tool
Description
Accepts transcript as string.
Returns chapters as text.
Parameters
Name | Type | Description |
---|
text | str | Transcript to be summarized into chapters. Assumes transcript contains time-stamped text. |
Description
Extracts JSON data from ingested file using LLM.
Accepts URL to be ingested, and JSON schema of Pydantic model to be extracted into. JSON schema needs be of type 'object' and include 'properties' and 'required' fields.
Returns extracted JSON from file.
Parameters
Name | Type | Description |
---|
uri | str | URL of cloud-hosted file to be ingested into knowledge base |
model_schema | str | Pydantic model JSON schema which describes the data which will be extracted. JSON schema needs be of type 'object' and include 'properties' and 'required' fields. |
prompt | Optional[str] | Text prompt which is provided to LLM to guide data extraction |
Description
Extracts JSON data from ingested web page using LLM.
Accepts URL to be scraped, and JSON schema of Pydantic model to be extracted into. JSON schema needs be of type 'object' and include 'properties' and 'required' fields.
Returns extracted JSON from web page.
Parameters
Name | Type | Description |
---|
uri | str | URL of web page to be scraped and ingested into knowledge base |
model_schema | str | Pydantic model JSON schema which describes the data which will be extracted. JSON schema needs be of type 'object' and include 'properties' and 'required' fields. |
prompt | Optional[str] | Text prompt which is provided to LLM to guide data extraction |
Description
Extracts JSON data from text using LLM.
Accepts text to be scraped, and JSON schema of Pydantic model to be extracted into. JSON schema needs be of type 'object' and include 'properties' and 'required' fields.
Returns extracted JSON from text.
Parameters
Name | Type | Description |
---|
text | str | Text to be extracted with LLM |
model_schema | str | Pydantic model JSON schema which describes the data which will be extracted. JSON schema needs be of type 'object' and include 'properties' and 'required' fields. |
prompt | Optional[str] | Text prompt which is provided to LLM to guide data extraction |
Support
Please refer to the Graphlit API Documentation.
For support with the Graphlit Agent Tools or to request an additional tool, please submit a GitHub Issue.
For further support with the Graphlit Platform, please join our Discord community.