Bundles

Documentation Index

Fetch the complete documentation index at: https://docs.llmcontrols.ai/llms.txt

Use this file to discover all available pages before exploring further.

Bundles are third-party components grouped by provider. For more information on bundled components, see the component provider’s documentation. The following components are available under Bundles.

• Apify - Integrate Apify with LLM Controls
• AssemblyAI - AssemblyAI components
• Astra DB - Connect an Astra DB MCP server to LLM Controls
• Composio - Integrate Composio with LLM Controls
• Google - Google service components
• Notion - Notion integration components
• NVIDIA - NVIDIA components

​

Agent components

Agents use LLMs as a brain to analyze problems and select external tools.

​

CrewAI Agent

This component represents an Agent of CrewAI allowing for the creation of specialized AI agents with defined roles goals, and capabilities within a crew. For more information, see the CrewAI documentation.

Parameters

Inputs

Name	Display Name	Info
role	Role	The role of the agent.
goal	Goal	The objective of the agent.
backstory	Backstory	The backstory of the agent.
tools	Tools	The tools at the agent’s disposal.
llm	Language Model	The language model that runs the agent.
memory	Memory	This determines whether the agent should have memory or not.
verbose	Verbose	This enables verbose output.
allow_delegation	Allow Delegation	This determines whether the agent is allowed to delegate tasks to other agents.
allow_code_execution	Allow Code Execution	This determines whether the agent is allowed to execute code.
kwargs	kwargs	Additional keyword arguments for the agent.

Outputs

Name	Display Name	Info
output	Agent	The constructed CrewAI Agent object.

​

Hierarchical Crew

This component represents a group of agents managing how they should collaborate and the tasks they should perform in a hierarchical structure. This component allows for the creation of a crew with a manager overseeing the task execution. For more information, see the CrewAI documentation.

Parameters

Inputs

Name	Display Name	Info
agents	Agents	The list of Agent objects representing the crew members.
tasks	Tasks	The list of HierarchicalTask objects representing the tasks to be executed.
manager_llm	Manager LLM	The language model for the manager agent.
manager_agent	Manager Agent	The specific agent to act as the manager.
verbose	Verbose	This enables verbose output for detailed logging.
memory	Memory	The memory configuration for the crew.
use_cache	Use Cache	This enables caching of results.
max_rpm	Max RPM	This sets the maximum requests per minute.
share_crew	Share Crew	This determines if the crew information is shared among agents.
function_calling_llm	Function Calling LLM	The language model for function calling.

Outputs

Name	Display Name	Info
crew	Crew	The constructed Crew object with hierarchical task execution.

​

CSV Agent

This component creates a CSV agent from a CSV file and an LLM.

Parameters

Inputs

Name	Type	Description
llm	LanguageModel	The language model to use for the agent.
path	File	The path to the CSV file.
agent_type	String	The type of agent to create.

Outputs

Name	Type	Description
agent	AgentExecutor	The CSV agent instance.

​

OpenAI Tools Agent

This component creates an OpenAI Tools Agent.

Parameters

Inputs

Name	Type	Description
llm	LanguageModel	The language model to use.
tools	List of Tools	The tools to give the agent access to.
system_prompt	String	The system prompt to provide context to the agent.
input_value	String	The user’s input to the agent.
memory	Memory	The memory for the agent to use for context persistence.
max_iterations	Integer	The maximum number of iterations to allow the agent to execute.
verbose	Boolean	This determines whether to print out the agent’s intermediate steps.
handle_parsing_errors	Boolean	This determines whether to handle parsing errors in the agent.

Outputs

Name	Type	Description
agent	AgentExecutor	The OpenAI Tools agent instance.
output	String	The output from executing the agent on the input.

​

OpenAPI Agent

This component creates an agent for interacting with OpenAPI services.

Parameters

Inputs

Name	Type	Description
llm	LanguageModel	The language model to use.
openapi_spec	String	The OpenAPI specification for the service.
base_url	String	The base URL for the API.
headers	Dict	The optional headers for API requests.
agent_executor_kwargs	Dict	The optional parameters for the agent executor.

Outputs

Name	Type	Description
agent	AgentExecutor	The OpenAPI agent instance.

​

Sequential Crew

This component represents a group of agents with tasks that are executed sequentially. This component allows for the creation of a crew that performs tasks in a specific order. For more information, see the CrewAI documentation.

Parameters

Inputs

Name	Display Name	Info
tasks	Tasks	The list of SequentialTask objects representing the tasks to be executed.
verbose	Verbose	This enables verbose output for detailed logging.
memory	Memory	The memory configuration for the crew.
use_cache	Use Cache	This enables caching of results.
max_rpm	Max RPM	This sets the maximum requests per minute.
share_crew	Share Crew	This determines if the crew information is shared among agents.
function_calling_llm	Function Calling LLM	The language model for function calling.

Outputs

Name	Display Name	Info
crew	Crew	The constructed Crew object with sequential task execution.

​

Sequential task agent

This component creates a CrewAI Task and its associated Agent, allowing for the definition of sequential tasks with specific agent roles and capabilities. For more information, see the CrewAI documentation.

Parameters

Inputs

Name	Display Name	Info
role	Role	The role of the agent.
goal	Goal	The objective of the agent.
backstory	Backstory	The backstory of the agent.
tools	Tools	The tools at the agent’s disposal.
llm	Language Model	The language model that runs the agent.
memory	Memory	This determines whether the agent should have memory or not.
verbose	Verbose	This enables verbose output.
allow_delegation	Allow Delegation	This determines whether the agent is allowed to delegate tasks to other agents.
allow_code_execution	Allow Code Execution	This determines whether the agent is allowed to execute code.
agent_kwargs	Agent kwargs	The additional kwargs for the agent.
task_description	Task Description	The descriptive text detailing the task’s purpose and execution.
expected_output	Expected Task Output	The clear definition of the expected task outcome.
async_execution	Async Execution	The boolean flag indicating asynchronous task execution.
previous_task	Previous Task	The previous task in the sequence for chaining.

Outputs

Name	Display Name	Info
task_output	Sequential Task	The list of SequentialTask objects representing the created tasks.

​

SQL Agent

This component creates an agent for interacting with SQL databases.

Parameters

Inputs

Name	Type	Description
llm	LanguageModel	The language model to use.
database	Database	The SQL database connection.
top_k	Integer	The number of results to return from a SELECT query.
use_tools	Boolean	This determines whether to use tools for query execution.
return_intermediate_steps	Boolean	This determines whether to return the agent’s intermediate steps.
max_iterations	Integer	The maximum number of iterations to run the agent.
max_execution_time	Integer	The maximum execution time in seconds.
early_stopping_method	String	The method to use for early stopping.
verbose	Boolean	This determines whether to print the agent’s thoughts.

Outputs

Name	Type	Description
agent	AgentExecutor	The SQL agent instance.

​

Tool Calling Agent

This component creates an agent for structured tool calling with various language models.

Parameters

Inputs

Name	Type	Description
llm	LanguageModel	The language model to use.
tools	List[Tool]	The list of tools available to the agent.
system_message	String	The system message to use for the agent.
return_intermediate_steps	Boolean	This determines whether to return the agent’s intermediate steps.
max_iterations	Integer	The maximum number of iterations to run the agent.
max_execution_time	Integer	The maximum execution time in seconds.
early_stopping_method	String	The method to use for early stopping.
verbose	Boolean	This determines whether to print the agent’s thoughts.

Outputs

Name	Type	Description
agent	AgentExecutor	The tool calling agent instance

​

XML Agent

This component creates an XML Agent using LangChain. The agent uses XML formatting for tool instructions to the Language Model.

Parameters

Inputs

Name	Type	Description
llm	LanguageModel	The language model to use for the agent.
user_prompt	String	The custom prompt template for the agent with XML formatting instructions.
tools	List[Tool]	The list of tools available to the agent.

Outputs

Name	Type	Description
agent	AgentExecutor	The XML Agent instance.

​

Agent QL (Extract Web Data)

This component extracts structured data from a web page using an AgentQL query or a Natural Language description. It supports configurable request modes, stealth mode, scrolling, and screenshots.

Parameters

Inputs

Name	Display Name	Info
api_key	API Key	Your AgentQL API key from dev.agentql.com
url	URL	The URL of the public web page you want to extract data from.
query	AgentQL Query	The AgentQL query to execute. Learn more at https://docs.agentql.com/agentql-query or use a prompt.
prompt	Prompt	A Natural Language description of the data to extract from the page. Alternative to AgentQL query.
is_stealth_mode_enabled	Enable Stealth Mode (Beta)	Enable experimental anti-bot evasion strategies. May not work for all websites at all times.
timeout	Timeout	Seconds to wait for a request.
mode	Request Mode	’standard’ uses deep data analysis, while ‘fast’ trades some depth of analysis for speed. Options: fast, standard.
wait_for	Wait For	Seconds to wait for the page to load before extracting data.
is_scroll_to_bottom_enabled	Enable scroll to bottom	Scroll to bottom of the page before extracting data.
is_screenshot_enabled	Enable screenshot	Take a screenshot before extracting data. Returned in ‘metadata’ as a Base64 string.

Outputs

Name	Display Name	Info
data	Data	Extracted structured data as a Data object.

​

Olivya (Place Call)

This component creates an outbound call request from Olivya’s platform. It allows configuring the agent’s phone number, recipient, introductory message, system prompt, and conversation history.

Parameters

Inputs

Name	Display Name	Info
api_key	API Key	Your API key for authentication
from_number	From Number	The Agent’s phone number
to_number	To Number	The recipient’s phone number
first_message	First Message	The Agent’s introductory message
system_prompt	System Prompt	The system prompt to guide the interaction
conversation_history	Conversation History	The summary of the conversation

Outputs

Name	Display Name	Info
output	Output	The call response as a Data object.

​

Cohere Rerank

This component reranks documents using the Cohere API. It takes search results and a query, then returns documents reranked by relevance using Cohere’s reranking models.

Parameters

Inputs

Name	Display Name	Info
search_query	Search Query	(inherited from base)
search_results	Search Results	Search Results from a Vector Store.
top_n	Top N	(inherited from base, default: 3)
api_key	Cohere API Key
model	Model	Options: rerank-english-v3.0, rerank-multilingual-v3.0, rerank-english-v2.0, rerank-multilingual-v2.0.

Outputs

Name	Display Name	Info
reranked_documents	Reranked Documents	Documents reranked by relevance

​

Firecrawl

Firecrawl is a suite of 4 components for web crawling, scraping, extracting, and mapping. All require a Firecrawl API Key.

Parameters

Inputs

Name	Display Name	Info
api_key	API Key	The API key to use Firecrawl API.
url	URL	The URL to scrape.
timeout	Timeout	Timeout in milliseconds for the request.
idempotency_key	Idempotency Key	Optional idempotency key to ensure unique requests.
crawlerOptions	Crawler Options	The crawler options to send with the request.
scrapeOptions	Scrape Options	The page options to send with the request.

Outputs

Name	Display Name	Info
data	Data	Crawl results as a Data object.

​

Firecrawl Scrape API

This component scrapes a single web page using the Firecrawl Scrape API.

Parameters

Inputs

Name	Display Name	Info
api_key	API Key	The API key to use Firecrawl API.
url	URL	The URL to scrape.
timeout	Timeout	Timeout in milliseconds for the request.
scrapeOptions	Scrape Options	The page options to send with the request.
extractorOptions	Extractor Options	The extractor options to send with the request.

Outputs

Name	Display Name	Info
data	Data	Scraped page data as a Data object.

​

Firecrawl Extract API

This component extracts structured data from web pages using the Firecrawl Extract API with prompt-guided extraction.

Parameters

Inputs

Name	Display Name	Info
api_key	API Key	The API key to use Firecrawl API.
url	URL	The URL to extract data from.
timeout	Timeout	Timeout in milliseconds for the request.
prompt	Prompt	Natural language prompt describing the data to extract.
schema	Schema	JSON schema defining the structure of data to extract.

Outputs

Name	Display Name	Info
data	Data	Extracted structured data as a Data object.

​

Firecrawl Map API

This component maps all URLs found on a website using the Firecrawl Map API.

Parameters

Inputs

Name	Display Name	Info
api_key	API Key	The API key to use Firecrawl API.
urls	URLs	List of URLs to create maps from (separated by commas or new lines).
ignore_sitemap	Ignore Sitemap	When true, the sitemap.xml file will be ignored during crawling.
sitemap_only	Sitemap Only	When true, only links found in the sitemap will be returned.
include_subdomains	Include Subdomains	When true, subdomains of the provided URL will also be scanned.

Outputs

Name	Display Name	Info
data	Data	Map results containing discovered URLs as a Data object.

​

Unstructured API

This component uses the Unstructured.io API to extract clean text from raw source documents. It supports a wide range of file types including PDF, DOCX, PPTX, XLSX, HTML, Markdown, images, and many more.

Parameters

Inputs

Name	Display Name	Info
path	Files	File upload input (inherited from BaseFileComponent).
file_path	Server File Path	Data object with a server file path property or a Message object with a path to the file (inherited from BaseFileComponent).
separator	Separator	Specify the separator to use between multiple outputs in Message format (inherited from BaseFileComponent).
silent_errors	Silent Errors	If true, errors will not raise an exception (inherited from BaseFileComponent).
delete_server_file_after_processing	Delete Server File After Processing	If true, the Server File Path will be deleted after processing (inherited from BaseFileComponent).
ignore_unsupported_extensions	Ignore Unsupported Extensions	If true, files with unsupported extensions will not be processed (inherited from BaseFileComponent).
ignore_unspecified_files	Ignore Unspecified Files	If true, Data with no server file path property will be ignored (inherited from BaseFileComponent).
api_key	Unstructured.io Serverless API Key	Unstructured API Key. Create at: https://app.unstructured.io/
api_url	Unstructured.io API URL	Unstructured API URL.
chunking_strategy	Chunking Strategy	Chunking strategy to use. Options: (empty), basic, by_title, by_page, by_similarity.
unstructured_args	Additional Arguments	Optional dictionary of additional arguments to the Loader. See https://docs.unstructured.io/api-reference/api-services/api-parameters for more information.

Outputs

Name	Display Name	Info
data	Data	Extracted document content as a list of Data objects.
dataframe	DataFrame	Extracted document content as a DataFrame object.
message	Message	Extracted document content as a Message object.

​

Git

The Git section contains 2 components: Git (loader) and GitExtractor.

Parameters

Inputs

Name	Display Name	Info
repo_source	Repository Source	Select whether to use a local repo path or clone from a remote URL. Options: Local, Remote.
repo_path	Local Repository Path	The local path to the existing Git repository (used if ‘Local’ is selected).
clone_url	Clone URL	The URL of the Git repository to clone (used if ‘Clone’ is selected).
branch	Branch	The branch to load files from. Defaults to ‘main’.
file_filter	File Filter	Patterns to filter files. For example: Include only .py files: ‘.py’. Exclude .py files: ‘!.py’. Multiple patterns can be separated by commas.
content_filter	Content Filter	A regex pattern to filter files based on their content.

Outputs

Name	Display Name	Info
data	Data	Loaded documents as a list of Data objects.

​

GitExtractor

This component analyzes a Git repository and returns file contents, directory structure, repository information, and statistics.

Parameters

Inputs

Name	Display Name	Info
repository_url	Repository URL	URL of the Git repository (e.g., https://github.com/username/repo)

Outputs

Name	Display Name	Info
text_based_file_contents	Text-Based File Contents	All text-based file contents as a Message (cropped to 300k characters).
directory_structure	Directory Structure	The repository’s directory tree as a Message.
repository_info	Repository Info	Repository metadata (name, URL, default branch, last commit, branches) as a list of Data objects.
statistics	Statistics	Repository statistics (total files, size, lines, binary files, directories) as a list of Data objects.
files_content	Files Content	Individual file contents with path and size as a list of Data objects.

​

Confluence

This component loads documents from a Confluence wiki collaboration platform using the Confluence API.

Parameters

Inputs

Name	Display Name	Info
url	Site URL	The base URL of the Confluence Space. Example: https://<company>.atlassian.net/wiki.
username	Username	Atlassian User E-mail. Example: email@example.com
api_key	API Key	Atlassian Key. Create at: https://id.atlassian.com/manage-profile/security/api-tokens
space_key	Space Key
cloud	Use Cloud?
content_format	Content Format	Specify content format. Options: editor, export_view, anonymous_export_view, storage, view. Defaults to storage.
max_pages	Max Pages	Maximum number of pages to retrieve in total, defaults 1000.

Outputs

Name	Display Name	Info
data	Data	Confluence page content as a list of Data objects.

​

YouTube

The YouTube section contains 7 components for interacting with YouTube data. YouTube Channel This component retrieves detailed information and statistics about YouTube channels as a DataFrame.

Parameters

Inputs

Name	Display Name	Info
channel_url	Channel URL or ID	The URL or ID of the YouTube channel.
api_key	YouTube API Key	Your YouTube Data API key.
include_statistics	Include Statistics	Include channel statistics (views, subscribers, videos).
include_branding	Include Branding	Include channel branding settings (banner, thumbnails).
include_playlists	Include Playlists	Include channel’s public playlists.

Outputs

Name	Display Name	Info
channel_df	Channel Info	Channel information as a DataFrame.

YouTube Comments This component retrieves and analyzes comments from YouTube videos.

Parameters

Inputs

Name	Display Name	Info
video_url	Video URL	The URL of the YouTube video to get comments from.
api_key	YouTube API Key	Your YouTube Data API key.
max_results	Max Results	The maximum number of comments to return.
sort_by	Sort By	Sort comments by time or relevance. Options: time, relevance.
include_replies	Include Replies	Whether to include replies to comments.
include_metrics	Include Metrics	Include metrics like like count and reply count.

Outputs

Name	Display Name	Info
comments	Comments	Video comments as a DataFrame.

YouTube Playlist This component extracts all video URLs from a YouTube playlist.

Parameters

Inputs

Name	Display Name	Info
playlist_url	Playlist URL	URL of the YouTube playlist.

Outputs

Name	Display Name	Info
video_urls	Video URLs	Playlist video URLs as a DataFrame.

YouTube Search This component searches YouTube videos based on a query.

Parameters

Inputs

Name	Display Name	Info
query	Search Query	The search query to look for on YouTube.
api_key	YouTube API Key	Your YouTube Data API key.
max_results	Max Results	The maximum number of results to return.
order	Sort Order	Sort order for the search results. Options: relevance, date, rating, title, viewCount.
include_metadata	Include Metadata	Include video metadata like description and statistics.

Outputs

Name	Display Name	Info
results	Search Results	Search results as a DataFrame.YouTube TrendingYouTube TrendingYouTube Trending

YouTube Trending This component retrieves trending videos from YouTube with filtering options by region and category.

Parameters

Inputs

Name	Display Name	Info
api_key	YouTube API Key	Your YouTube Data API key.
region	Region	The region to get trending videos from. Options: Global, United States, Brazil, United Kingdom, India, Japan, South Korea, Germany, France, Canada, Australia, Spain, Italy, Mexico, Russia, Netherlands, Poland, Argentina.
category	Category	The category of videos to retrieve. Options: All, Film & Animation, Autos & Vehicles, Music, Pets & Animals, Sports, Travel & Events, Gaming, People & Blogs, Comedy, Entertainment, News & Politics, Education, Science & Technology, Nonprofits & Activism.
max_results	Max Results	Maximum number of trending videos to return (1-50).
include_statistics	Include Statistics	Include video statistics (views, likes, comments).
include_content_details	Include Content Details	Include video duration and quality info.
include_thumbnails	Include Thumbnails	Include video thumbnail URLs.

Outputs

Name	Display Name	Info
trending_videos	Trending Videos	Trending videos as a DataFrame.

YouTube Video Details This component retrieves detailed information and statistics about YouTube videos.

Parameters

Inputs

Name	Display Name	Info
video_url	Video URL	The URL of the YouTube video.
api_key	YouTube API Key	Your YouTube Data API key.
include_statistics	Include Statistics	Include video statistics (views, likes, comments).
include_content_details	Include Content Details	Include video duration, quality, and age restriction info.
include_tags	Include Tags	Include video tags and keywords.
include_thumbnails	Include Thumbnails	Include video thumbnail URLs in different resolutions.

Outputs

Name	Display Name	Info
video_data	Video Data	Video details as a DataFrame.

YouTube Transcripts This component extracts spoken content from YouTube videos with multiple output options.

Parameters

Inputs

Name	Display Name	Info
url	Video URL	Enter the YouTube video URL to get transcripts from.
chunk_size_seconds	Chunk Size (seconds)	The size of each transcript chunk in seconds.
translation	Translation Language	Translate the transcripts to the specified language. Options: (empty), en, es, fr, de, it, pt, ru, ja, ko, hi, ar, id. Leave empty for no translation.

Outputs

Name	Display Name	Info
dataframe	Chunks	Transcript chunks with timestamps as a DataFrame.
message	Transcript	Full transcript as continuous text in a Message object.
data_output	Transcript + Source	Transcript text with video URL metadata as a Data object.

​

ScrapeGraph AI

ScrapeGraph AI is a suite of 3 components for web scraping and search using the ScrapeGraph API. ScrapeGraph Markdownify API
This component converts a web page into Markdown content using the ScrapeGraph Markdownify API.

Parameters

Inputs

Name	Display Name	Info
api_key	ScrapeGraph API Key	The API key to use ScrapeGraph API.
url	URL	The URL to markdownify.

Outputs

Name	Display Name	Info
data	Data	Markdownified content as a Data object.

ScrapeGraph Search API This component performs search queries using ScrapeGraph’s search functionality and returns structured search results.

Parameters

Inputs

Name	Display Name	Info
api_key	ScrapeGraph API Key	The API key to use ScrapeGraph API.
user_prompt	Search Prompt	The search prompt to use.

Outputs

Name	Display Name	Info
data	Data	Search results as a Data object.

ScrapeGraph Smart Scraper API This component scrapes a website and returns structured data using ScrapeGraph’s AI-powered Smart Scraper.

Parameters

Inputs

Name	Display Name	Info
api_key	ScrapeGraph API Key	The API key to use ScrapeGraph API.
url	URL	The URL to scrape.
prompt	Prompt	The prompt to use for the smart scraper.

Outputs

Name	Display Name	Info
data	Data	Structured scraped data as a Data object