Chat Message: Difference between revisions

From Tygron Support wiki
Jump to navigation Jump to search
No edit summary
No edit summary
Line 2: Line 2:


Chat Messages are the content of [[Chat Channel]]s. Each channel can have multiple messages. A message may be sent by a User or the response of an [[AI Agent]].
Chat Messages are the content of [[Chat Channel]]s. Each channel can have multiple messages. A message may be sent by a User or the response of an [[AI Agent]].
Chat messages have the following properties:
* Message; The answer.
* Thinking; The thinking and reasoning process of an Agent (if enabled).
* [[Agent Tool]] Calls.
* [[#Agent State|Agent State]].
* [[#Role|Role]]. The entity that send the message.
* [[Item ID]] of the related [[Chat Channel]]
* The time it took to generate the message, in milliseconds.
* The date the message was generated, in milliseconds epoch.


==Role==
==Role==

Revision as of 13:37, 9 October 2026

This functionality is currently in BETA.
Read more about Beta features.


Chat Messages are the content of Chat Channels. Each channel can have multiple messages. A message may be sent by a User or the response of an AI Agent.

Chat messages have the following properties:

  • Message; The answer.
  • Thinking; The thinking and reasoning process of an Agent (if enabled).
  • Agent Tool Calls.
  • Agent State.
  • Role. The entity that send the message.
  • Item ID of the related Chat Channel
  • The time it took to generate the message, in milliseconds.
  • The date the message was generated, in milliseconds epoch.

Role

There are several message role types:

  • SYSTEM: System role is used to provide the AI Agent instructions, in the first message.
  • INTRO: Intro or welcome message, not processed by AI Agent.
  • ASSISTANT: Messages generated by the AI Agent.
  • USER: Messages typed by the User.

Agent State

There are several message sates that describe AI Agent activity:

  • NONE: Message has not been processed yet.
  • QUEUING: Message queued for processing by AI Agent.
  • READING: AI Agent is reading the prompt.
  • GENERATING: AI Agent is generating the contents.
  • FINISHED: AI Agent has finished.

Notes

  • To properly use this technology and to protect your data, several measures have been taken. These are described in our AI Rules.

See also

API Endpoints