# Agents - [What is an agent?](https://docs.katara.ai/about/agents/what-is-an-agent.md): Agents are reusable units that perform tasks like loading documentation data or connecting Katara to community platforms, and can be linked to form workflows. - [Loaders](https://docs.katara.ai/about/agents/loaders.md): Loaders in Katara 1.0 ingest data from external sources, capturing key content while ignoring images and code. They auto-refresh and update the DocDB. - [Gitbook](https://docs.katara.ai/about/agents/loaders/gitbook.md): This loader extracts text data from GitBook, skipping images and code blocks, with auto-refresh and change alerts. Configurable URL base, refresh rate, and exclusions. - [Website](https://docs.katara.ai/about/agents/loaders/website.md): This loader extracts raw text from websites, ignoring images and code blocks, with refresh and change tracking. Supports URL or sitemap scraping; future media parsing planned. - [GitHub](https://docs.katara.ai/about/agents/loaders/github.md): This loader pulls Markdown and text files from GitHub, auto-refreshing and tracking changes. It notifies users of updates and supports one agent per branch. - [Notion](https://docs.katara.ai/about/agents/loaders/notion.md): This loader collects data from Notion pages. - [Content Uploader](https://docs.katara.ai/about/agents/loaders/content-uploader.md): This loader allows you to upload single files directly into Katara. - [AI Insights](https://docs.katara.ai/about/agents/ai-insights.md): Generative agents handle advanced AI tasks like content creation and data analysis, using data from loaders to generate responses based on prompts or rules. - [Q\&A](https://docs.katara.ai/about/agents/ai-insights/q-and-a.md): Q\&A agents generate answers with summaries, citations, and confidence scores. They link to platforms like Discord and use confidence thresholds to ensure quality. - [Taxonomy](https://docs.katara.ai/about/agents/ai-insights/taxonomy.md): The taxonomy agent creates topic lists with keywords from your data, updated via manual, scheduled, or re-training runs for use in classification. - [Conversational](https://docs.katara.ai/about/agents/conversational.md): Social agents connect Katara to platforms like Discord, Telegram, and Slack, enabling community Q\&A and data collection for analysis and support workflows. - [Discord](https://docs.katara.ai/about/agents/conversational/discord.md): The Discord agent links Katara to your server, enabling message handling, data collection, and Q\&A replies with configurable permissions and settings. - [Ignore Messages From Some Users](https://docs.katara.ai/about/agents/conversational/discord/ignore-messages-from-some-users.md) - [Telegram](https://docs.katara.ai/about/agents/conversational/telegram.md): The Telegram agent links Katara to your group, enabling message handling, data collection, and configurable Q\&A replies via commands and reply policies. - [Slack](https://docs.katara.ai/about/agents/conversational/slack.md): The Slack agent links Katara to your workspace, enabling message handling, data collection, and configurable Q\&A replies with commands and reply policies. - [Website Widget](https://docs.katara.ai/about/agents/conversational/website-widget.md): Embed a customizable Katara support bot on your website with a generated script tag. Control widget look, responses, and branding via HTML data attributes and CSS. - [Classifiers](https://docs.katara.ai/about/agents/classifiers.md): Classifier agents tag documents by categories like Topic, Language, or Sentiment. They run on schedules, with configurable filters, thresholds, and report options. - [Topic Classifier](https://docs.katara.ai/about/agents/classifiers/topic-classifier.md): The topic classifier tags data using topics from a taxonomy. - [Flesch Reading Ease Classifier](https://docs.katara.ai/about/agents/classifiers/flesch-reading-ease-classifier.md): The Flesch Reading Ease Classifier rates how easy your data is to read using the Flesch formula. - [Language Classifier](https://docs.katara.ai/about/agents/classifiers/language-classifier.md): The Language Classifier classifies your data based on what language it is. - [Passive Voice Classifier](https://docs.katara.ai/about/agents/classifiers/passive-voice-classifier.md): The Passive Voice Classifier classifies your data based on the amount of passive voice it uses. - [Sentiment Classifier](https://docs.katara.ai/about/agents/classifiers/sentiment-classifier.md): The Sentiment Classifier classifies data by sentiment: positive, neutral, or negative. - [General LLM Classifier](https://docs.katara.ai/about/agents/classifiers/general-llm-classifier.md): The General LLM Classifier classifies your data into topics that you define. - [Diátaxis Classifier](https://docs.katara.ai/about/agents/classifiers/diataxis-classifier.md): The Diátaxis Classifier classifies your documentation into the four categories of the Diátaxis framework: tutorials, how-to guides, reference material, and explanation. - [Tone Classifier](https://docs.katara.ai/about/agents/classifiers/tone-classifier.md): The Tone Classifier classifies your data based on the writing tone used in each document. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.katara.ai/about/agents.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.