# Joggr

> Joggr builds a knowledge graph that connects code, conversations, tickets, and documentation. AI coding tools access this graph through MCP to get accurate context.

## What Joggr Does

Joggr automatically captures engineering knowledge from multiple sources and delivers it in two forms:
- Human-readable documentation for developers (also accessible to AI agents via filesystem)
- prebuilt & structured context for AI agents via MCP (Model Context Protocol)

## Key Features

- **Context Without Limits**: Delivers exactly what AI needs, no token bloat, no context rot
- **Works Everywhere**: Native MCP integration with Claude Code, Cursor, Windsurf, Copilot, and more
- **Always Up-to-Date**: Real-time sync across code, conversations, and tickets
- **Zero Maintenance**: Auto-generates missing docs and auto-fixes outdated docs on PRs
- **Extensible Platform**: Build custom Slackbots, agents, and workflows via MCP and APIs
- **Universal Context**: Search across GitHub, Slack, Jira, Linear, Confluence, and more

## How It Works

1. **Connect and Generate**: Install on GitHub, connect tools (Slack, Jira, Linear). Joggr analyzes code, conversations, and tickets to build your knowledge base automatically
2. **Always Up-to-Date**: As code changes, Joggr maintains documentation in real-time
3. **Supercharge Agents**: Provides exact context AI agents need for accurate outputs and lower token usage

## Integrations

- **AI Coding Agents**: Claude Code, Cursor, Windsurf, GitHub Copilot, Devin, Gemini, Augment Code, CodeRabbit
- **Code Platforms**: GitHub, GitLab
- **Collaboration**: Slack, Microsoft Teams, Confluence, Notion, Jira, Linear, Miro

## Problems Solved

- Context rot degrading LLM accuracy (up to 50% with distractors)
- Token waste from repeated codebase searches
- Scattered knowledge across wikis, READMEs, Slack
- Stale documentation that diverges from code
- Manual documentation burden on developers

## Results

- Up to 80% more accurate AI responses
- 2-3x faster task-to-completion time
- 30% fewer errors