# GPA

Slate Group Processing Agent

### Overview

The **Slate GPA** is a specialized Telegram bot that combines selected users IPAs, finds alpha across all the IPAs and sends a live feed of alerts into a designated "destination" group chat for viewing.

#### Why use the GPA?

* **Frictionless Alpha Sharing:** Automatically distribute valuable token calls from your IPA’s monitored group chats directly to your closest friends, streamlining alpha sharing effortlessly.
* **Comprehensive Coverage:** Gain access to crucial alpha even from group chats you're not personally in—but your friends are—ensuring you never miss valuable opportunities. Your Slate IPA becomes more useful if you and your friends are all tracking separate groups.
* **Higher Signal Quality:** All alerts are cross-referenced with a vastly expanded set of chats (from both you and your friends), significantly enhancing the reliability and confidence of notifications.

### 1. Onboarding

The GPA is currently a **white-glove service**. You will need to contact the team directly to onboard.

1. Find **group of 5+ friends** and have everyone onboard to the [IPA](/products/ipa.md).
2. Have each of your friends [add chats to track](https://docs.slate.ceo/products/pages/kvgp8XFngRmjNnlitrkH#id-3.-essential-commands) within their individual IPAs.
3. Decide where (what Telegram channel) you want the GPA to message in.
4. Contact **@shirtlessfounder** on Telegram to request access to the GPA. Include the Telegram handles of all your onboarded friends who wish to participate.
5. Once approved, add @shirtlessfounder to the channel in Step 3.
6. Let us finish up the set up process.

### 2. How the GPA Works

**GPA** listens to all of the group chats you and your group are tracking via your IPAs. Whenever someone calls mentions a token ticker (`$XYZ`) or a contract address (`0x...`) in these groups, the bot notes:

* **Which user** mentioned it
* **Which group** it was mentioned in
* **Timestamp** of mention

It accumulates mentions in a short-term “mention tracker.” Then:

* **Confluence Threshold:** If the same token is mentioned by at least **N** unique callers within a rolling time window (3 hours by default).
* **CABAL Alert:** If a token is mentioned by **3+ unique users** but **only in a single group**, it triggers a “CABAL” alert (suggesting coordinated posting in one group).
* **Repeat Alerts:** If an alert has already been sent for a token, the bot requires more unique callers than before to trigger a new alert (so it doesn’t spam on the same mention count).

### 3. Understanding the Alerts

When the GPA detects a confluence or cabal scenario, it sends a message with a format like:

```
🥇 CONFL 🥇 [3c | 5m | 2g] on $FOO [MC: $2.4M]

🗣️ Callers: @userA [2], @userB [TBD], ...
🏘️ Groups: ApesChat [3m], DeFiAlpha [2m]
🔎 IPAs: ...

🔗 DS | X | Pump | ...
____________________
```

[Breaking this down (reference IPA notifications):](https://docs.slate.ceo/products/pages/kvgp8XFngRmjNnlitrkH#id-5.-understanding-the-alerts)

* Header**:** `🥇 CONFL 🥇` or `♻️ CONFL ♻️` or `⚠️ CABAL ⚠️`
* \[3c | 5m | 2g]**:**
  * `3c` = 3 unique callers
  * `5m` = 5 total mentions
  * `2g` = across 2 groups
* $FOO \[MC: $2.4M]**:** The token symbol `$FOO` and approximate market cap if retrieved from DexScreener or other sources.
* **🗣️** Callers**:** Lists top 3 callers, plus their caller ELO in brackets. \[2] means that user’s ELO is 2, `[TBD]` means they have no recent track record.
* **🏘️** Groups**:** Shows which groups posted about `$FOO`, with mention counts.
* **🔎 IPAs: Shows which personal IPA bots or user sessions are also tracking that group.**
* **🔗** (Links): Quick links to DexScreener, Twitter, Pump, etc. for deeper investigation.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.slate.ceo/products/gpa.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.