# AI Agent persistent conversation IDs

## **Persistent AI Agent Conversations**

Create long-running conversations around a single record, task or topic by giving the conversation a unique thread ID.\
Multiple triggers can then send events into that same conversation thread, allowing your agent to responds to each event with all the context and plans it had previously.\
Use the thread in detail page widgets to give access to a persistent conversation, scoped to a particular record, user, or anything else you want.

### How to set up on detail pages: 

1. Add an AI agent widget to your detail page
2. Set your template ID to keep it persistent based on record ID, user ID, or something else:
   1. **Record ID** use `{record.id}`: this will keep the conversation the same based on the record regardless of the user logged in. 
   2. **User Record** use `{record.user}`: this will keep the conversations or other AI tasks based on the user logged in. 
   3. Other fields use `{record.customer.primary_contact.email}`
      1. The path values here correspond to what is shown for the ‘developer name’ in the Manage Fields modal.
      2. If it is a multi-link, we will use only the first record in the list.

{% embed url="<https://vimeo.com/1093226590?share=copy>" %}

### Tool usage for Persistent IDs:

Use the **Record Creation** trigger along with a Conversation ID set to connect with specific users or other conversation records.

For example, when adding new records that associate with the same company, using the {record.company} will help the AI agent reference all record information that is in common with the company. Helping your AI agent gather more accurate and complete information.&#x20;

<figure><img src="/files/ZoqFMdZQCjOHwWKN3reh" alt=""><figcaption></figcaption></figure>


---

# 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.stackerhq.com/stacker-astra/ai-and-automations/ai-agents-beta/setting-up-your-ai-agent/ai-agent-persistent-conversation-ids.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.