Capabilities
These capabilities are available in the ADITO platform. Each section describes what is possible today and links to the relevant API feature.
What you can build today
🧩 Structured AI outputs for automation
AI responses can follow a defined JSON schema so they flow directly into automated workflows or get imported as CRM entities. No manual parsing required.
Use the guided_json parameter when calling the ADITO-LLM API to enforce a response schema.
💼 CRM-aware prompts
Include CRM data (accounts, opportunities, activities) in prompts. Common use cases: summarizing a customer's history, suggesting next actions, drafting emails from existing records.
Include CRM data in the messages array when calling the Chat Completions endpoint. For example, pass an account summary in the system message and the user's question in the user message.
💬 Natural-language task execution
Describe what you need in plain language. A prompt like "Create a quote based on this customer and their last opportunities" produces a draft you can review and apply.
Combine CRM-aware prompts with function calling to let the model invoke your web services based on natural-language instructions.
🏷️ Automatic classification
Emails, notes, and support tickets can be sorted automatically by priority, topic, or responsible team. The model processes unstructured text and assigns categories based on criteria you define.
Use guided_json to constrain the model output to your classification schema (e.g., an enum of priority levels or topic labels).
🔄 Scheduled record enrichment
Server-side jobs can call the AI on a schedule to enrich records with summaries, keywords, or classifications. This keeps data up to date without manual effort.
Call the ADITO-LLM API from a scheduled server process or ADITO web service.
🌐 Web scraping for context
External information can be pulled from the web and fed into AI prompts. An MCP-connected tool or ADITO web service fetches the page content, which is then included in the prompt as additional context. This is useful when the CRM alone does not contain enough information to produce a good answer.
Configure an MCP server or a custom web service to retrieve external content, then pass it alongside CRM data in the messages array.
See the ADITO-LLM API reference for endpoint details and code examples. For planned features, see the Roadmap.