Data Sources

Documentation Index

Fetch the complete documentation index at: https://docs.dialog.tools/llms.txt

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

Dialog connects to multiple specialized data sources to give your agent comprehensive research coverage. Each source is optimized for a different type of intelligence gathering.

​

Reddit Research

Enabled by default for all agents. Reddit Research gives your agent deep access to community discussions across the platform.

• Discover subreddits relevant to any topic or industry
• Fetch posts and comments with full thread context
• Analyze sentiment across discussions and communities
• Semantic search to find conversations by meaning, not keywords

Reddit is Dialog’s foundational data source — it powers everything from competitive intelligence to customer discovery and pain point analysis.

​

Discourse Forums

Search and read content from Discourse-powered community forums and support sites.

• Search forum topics across categories and tags
• Read full posts and replies with author context
• Browse categories to understand community structure

Discourse forums are common for open-source projects, SaaS products, and developer communities, making this source valuable for technical research and product feedback.

​

Web Search & Scraping

Search the open web and extract content from any public page.

• Search the web using natural language queries
• Extract page content from specific URLs
• Crawl websites to gather content across multiple pages
• Extract structured data from pages into organized formats

This source works well alongside Reddit and social media — use it to pull official documentation, blog posts, news articles, or product pages into your research.

​

Social Media Intelligence

Access posts, profiles, and engagement data across major social platforms.

​

Instagram

Profiles, posts, reels, and comments. Useful for brand monitoring and influencer research.

​

Facebook

Posts and company reviews. Good for understanding public brand perception and customer feedback.

​

TikTok

Profiles, posts, and comments. Helpful for tracking trends and content performance in short-form video.

​

YouTube

Profiles, videos, and comments. Valuable for product reviews, tutorials, and audience sentiment analysis.

​

X (Twitter)

Posts and public discussions. Strong for real-time sentiment, industry conversations, and thought leader tracking.

​

Business Intelligence

Pull company data from professional databases. Useful for competitive analysis, market mapping, and lead research when you need structured company or professional profile information.

​

App Store Data

Access reviews and listings from the Google Play Store and Apple App Store. This source is especially useful for product teams tracking user feedback, monitoring competitor app ratings, and identifying feature requests.

​

Connected Apps

Connect your own tools to Dialog via OAuth for additional agent capabilities. Available integrations include:

• Communication: Gmail, Slack, Zoom
• Project Management: Jira, Linear, Asana, Notion
• Productivity: Google Calendar, Google Sheets, Google Docs
• CRM & Sales: HubSpot, Salesforce, Intercom
• Analytics & Product: Google Analytics, PostHog, Productboard
• Design & Development: Figma, GitHub, LinkedIn

Connect apps in Settings — each integration uses OAuth so your credentials stay secure. Once connected, your agent can read and interact with these services during research sessions.

​

How sources work together

The real power of Dialog’s data sources shows up when your agent combines multiple sources in a single query. Ask something like “What do developers think about Cursor vs Windsurf?” and your agent might search Reddit for community discussions, pull X posts for real-time opinions, check Discourse forums for detailed technical threads, and scrape comparison blog posts from the web — all in one pass. The agent decides which sources to query based on your question and your tool preferences. Results from different sources are synthesized into a single research brief with citations pointing back to each original source.