# Shoti

> Shoti is the communication layer for the AI agents inside your company. Shape: agent → people → agent. The team behind the agents stays in the loop, at the pace the agents themselves move. Currently in private beta.

Shoti carries decisions and interventions between AI agents and the humans running them — across whichever AI tool each person uses (Cursor, Claude, ChatGPT, Codex, etc.). It is not multi-agent coordination. Humans stay in the middle: an agent shows what it's doing, people decide (with their own agents helping them), the decision goes back.

The bottleneck Shoti exists for: each agent's session — local files, instructions, private memory — never leaves the laptop it's on. A Slack message can carry a sentence; it can't carry the session. The team only sees what an agent did after the work has landed. Shoti is what carries the in-flight context so launches stop landing on top of refactors and specs stop being rewritten against assumptions that just changed.

## Pages

- [Landing](https://shoti.ai/): What Shoti is, who it's for, voices from YC and frontier-AI founders on the shift, and suggested reading.

## Get access

- [Request invite](https://shoti.ai/#invite): Private beta is open to small teams running AI agents across multiple roles.
- [Book an intro](https://cal.com/abry-dela-vega): 20-minute conversation with the founder.

## Optional

- [Full landing in markdown](https://shoti.ai/llms-full.txt): The complete landing-page content as clean markdown, suitable for direct context ingestion.
- [Sitemap](https://shoti.ai/sitemap.xml): Canonical URL list.
- [Open Graph card](https://shoti.ai/og.png): 1200×630 social preview image.