# Celync

> Celync builds agents and agentic workflows that make it easier than ever to get the right data into the right places — for both humans and AI assistants. Pipelines, connectors, monitoring, and certification are the surfaces those agents act on; humans approve and own the result.

Celync is an agentic data engineering platform. AI agents draft a pipeline from a plain-language description, a human approves it, and the platform runs, monitors, and self-heals it. The product is delivered as a managed SaaS today and is architected for on-prem / VPC deployment.

The product exposes two MCP entry points (Forge and Tap) on top of the same managed core (HTTP API + engine + data plane) that powers the web app. MCP is how AI assistants reach Celync; it is not a separate product.

This `llms.txt` follows the proposal at https://llmstxt.org/. Markdown documents below are stable, agent-friendly summaries of what Celync is and how it works.

## Core

- [About Celync](/about.md): What Celync is, who it's for, mission.
- [Product offering](/product.md): Phase 0 capabilities and what ships together.
- [Agents and agentic workflows](/agents.md): How agents drive the platform end to end.
- [Connector architecture](/connectors.md): Built-in registry connectors plus agent-generated connector packages.
- [MCP entry points: Forge and Tap](/mcp.md): How AI hosts (Claude, Cursor, internal copilots) reach Celync.
- [FAQ](/faq.md): Concise answers for common questions.

## Optional

- [Privacy](/privacy): Cookie and analytics policy for celync.com.