AI Calls & Bots - API Documentation
Pre-configured AI calls for custom prompts, server monitoring, website monitoring, and log analysis.
In all URLs below, replace XXX with your AI Apps Account ID, found in your account settings.
What Are AI Bots?
Overview of bot types and how they fit into the platform
AI Bots is the app for making custom AI calls. Each bot is a saved configuration with a specific AI model, custom rules, and an optional extra prompt. When you run a bot, it sends your prompt to the configured AI model with all of your rules applied and returns the response. Bots can be called directly through the API, or as a step inside a Chain Command to add AI processing to any automation sequence.
The main purpose of this app is to give you reusable, pre-configured AI calls. Instead of writing AI integration code yourself or repeating the same rules every time, you set up a bot once and call it by ID whenever you need it. The general bot type handles any kind of AI request, making it the go-to choice for custom AI calls in your workflows. The monitoring bot types are purpose-built for specific automated tasks like checking server health, website status, or log files.
Bots support both ChatGPT and Claude models, including custom fine-tuned models you have trained through the Fine Tuning app. The model is selected per bot, so you can have different bots using different AI providers for different tasks.
Bot Types
The four types of AI bots and what each one does
General
A general-purpose AI call. You send a prompt, and the bot processes it with whatever rules you have configured. This is the type to use for any custom AI task, whether it is generating content, answering questions, analyzing text, classifying data, or anything else you need AI for. When used inside a Chain Command, the general bot becomes a powerful building block for adding AI reasoning to any step in your automation.
A general-purpose AI call. You send a prompt, and the bot processes it with whatever rules you have configured. This is the type to use for any custom AI task, whether it is generating content, answering questions, analyzing text, classifying data, or anything else you need AI for. When used inside a Chain Command, the general bot becomes a powerful building block for adding AI reasoning to any step in your automation.
Linux Server Monitor
Analyzes the output of Linux diagnostic commands (such as top, df, free, systemctl, or any other command output) and returns a single-word status:
Analyzes the output of Linux diagnostic commands (such as top, df, free, systemctl, or any other command output) and returns a single-word status:
ok, ignore, warning, alert, restart, or critical. Designed to be called automatically by a monitoring script that collects server output and posts it to this bot for AI analysis. The single-word response makes it easy to trigger different actions based on the result.
Website Monitor
Fetches the content of a URL and analyzes it, returning a single-word status:
Fetches the content of a URL and analyzes it, returning a single-word status:
ok, down, error, alert, or ignore. You provide the URL and the bot handles fetching the page content. Use this to monitor whether a website is loading normally, returning errors, or showing unexpected content.
Log Monitor
Analyzes the contents of a log file and returns a single-word status:
Analyzes the contents of a log file and returns a single-word status:
ok, ignore, warning, alert, or critical. Send the log content as the filecontent field and the bot determines whether the entries are normal, show unusual patterns, or indicate failures requiring attention.
All monitoring bot types return a single word so the response can easily drive automated decisions. In a Chain Command, you can use a conditional or switch node on the bot's response to take different actions, for example sending an alert email only when the status is "critical" or "alert".
Admin: Managing Bots
Creating and configuring your AI bots
The bots list page shows all AI bots on your account, grouped by bot type. From here you can create new bots and edit existing ones. Each bot is an independent configuration with its own model, rules, and settings.
Bot Settings
| Field | Description |
|---|---|
| Name | A display name for this bot. Used to identify it in the admin panel. |
| Bot Type | The type of AI call this bot makes: general, linuxmonitor, websitemonitor, or logmonitor. Each type has its own built-in behavior and expected input fields. |
| AI Model | Which AI model to use. Supports ChatGPT models, Claude models, and custom fine-tuned models. Each bot can use a different model. |
| Custom Rules | Optional rules that are added to the bot's system prompt. For general bots, these define how the AI should behave. For monitoring bots, these are added on top of the built-in monitoring rules. |
| Extra Prompt | Optional text that is appended to every prompt sent to this bot. Use this for context or instructions that should always be included. |
| Notes | Optional internal notes. Not used in execution. |
How Endpoints Work
Two ways to call each endpoint - read this first
App Command (API Call)
POST your request to
Mode / Webhook (Direct URL)
POST directly to the endpoint URL shown in each section. The fields go in the POST body as standard form fields (the same fields you would include in an HTML form POST). Use this method for provider webhooks, website forms, or any third-party system that sends form data. No API key is required, the account ID is part of the URL.
POST your request to
https://api.aiappsapi.com with your API key and account ID in the POST body. The fields described in each section below go inside the jsonData POST field as a JSON-encoded string. Use this method when your server is making the call and you have your API credentials available.Mode / Webhook (Direct URL)
POST directly to the endpoint URL shown in each section. The fields go in the POST body as standard form fields (the same fields you would include in an HTML form POST). Use this method for provider webhooks, website forms, or any third-party system that sends form data. No API key is required, the account ID is part of the URL.
Run Bot
command: runbot
Runs an AI bot and returns the response. The bot's type determines which fields are required and how the prompt is processed. General bots require a prompt. Linux and log monitor bots require filecontent. Website monitor bots require a URL. The response contains the AI model's output in the message field.
POST App Command URL
https://api.aiappsapi.com app: aibots command: runbot
This command is only available as an App Command. It is not available as a direct mode/webhook URL.
Required Fields
| Field Name | Field Key | Notes |
|---|---|---|
| Bot ID | botID | The index of the bot in your bots list. |
Fields by Bot Type
| Field Name | Field Key | Notes |
|---|---|---|
| Prompt | prompt | The text prompt to send to the AI. Required for general bots. |
| File Content | filecontent | The text content to analyze. Required for linuxmonitor and logmonitor bots. Send the raw output of server commands or log file contents as a string. |
| URL | url | The URL to check. Required for websitemonitor bots. The bot fetches the page content automatically. |