> ## Documentation Index
> Fetch the complete documentation index at: https://learn.social.plus/llms.txt
> Use this file to discover all available pages before exploring further.

# Agentry

> Ask questions about your first-party data in plain language — in ChatGPT, Claude, and other AI apps.

Agentry is how brands use social.plus in the AI-native era. Every post, comment, and reaction your app generates is first-party data — a signal about what your users care about and how they behave. Agentry puts that data to work where your team already is: ask in plain English from ChatGPT, Claude, or any AI app, and get back a clear summary plus the underlying data, ready to tabulate, chart, or drill into with follow-up questions.

<Frame>
  <img src="https://mintcdn.com/social-b97141fb/-t88VJFqwt-Dt9V9/images/ai/agentry-demo.svg?fit=max&auto=format&n=-t88VJFqwt-Dt9V9&q=85&s=0296da4dc96ba133b3f8e4352f3cc9e0" alt="Agentry answering 'How did engagement change this month?' with a summary and weekly engagement chart" width="760" height="400" data-path="images/ai/agentry-demo.svg" />
</Frame>

<Info>
  Agentry is in **preview**. If the Agentry page in your Console says it isn't enabled, contact support to turn it on.
</Info>

## What you can ask

Agentry reasons over your network's engagement activity — posts, comments, reactions, stories, events, users, and communities — as counts, trends, comparisons, cohorts, and rankings across roughly the last 90 days. If a question combines those ingredients — *what happened, measured how, over which period, compared to what* — Agentry can usually answer it.

Simple questions work — *how many users were active yesterday?* — and so do the analyses your team actually runs:

* **Campaign readouts** — *Did engagement lift after last week's launch?* See a campaign's impact the moment you ask.
* **Audience understanding** — *Which communities and member cohorts grew the most this month?* Know where growth comes from.
* **Content strategy** — *Which content formats drive the most engagement?* Double down on what works.
* **Community health** — *How is activity trending versus last month, and where is conversation concentrating?* Walk into the review with numbers in hand.
* **Trendspotting** — *Which communities saw the biggest spike in conversation this week?* Catch momentum while it's still building.
* **Projections** — *At the current pace, how many active users will we end the month with?* Simple forward reads from the trends in your data.

Answers arrive ready for your AI app to work with — it can chart the numbers, explain what they mean, and dig into the details when you ask a follow-up.

Out of scope, by design: activity older than the data window, anything outside your own network, and acting on your app — Agentry answers questions; it doesn't post, moderate, or administer.

## How it works

* **Read-only by design.** Agentry answers questions by reading your analytics data. It never posts, modifies, or deletes anything in your app.
* **Your data stays yours.** Each connection is scoped to your own network's data and authorized by your administrator from the Console. Admins see every connection and can remove any of them at any time.
* **Fresh, bounded data.** Activity data covers roughly the last 90 days and refreshes daily.
* **Self-contained questions.** Each question is answered on its own, so include the context that matters — time ranges, baselines, comparisons. AI apps handle this naturally when you ask follow-ups in the same conversation.

## Get started

<CardGroup cols={2}>
  <Card title="Connect your AI app" icon="plug" href="/ai/agentry/connect">
    Link ChatGPT, Claude, Claude Code, Cursor, or any MCP-capable app to Agentry.
  </Card>

  <Card title="Manage access" icon="key" href="/ai/agentry/manage-access">
    Generate connect codes, issue credentials, and control who can ask — from the Console.
  </Card>
</CardGroup>
