- Perception: Gather information as much as possible:
- From prompt via user
- From predefined information regarding to user
- From logs
- From APIs
- From other agents (MCP, A2A etc)
- Processing & Decision Making:
- Choose LLM to use
- Choose the best prompt for the problem
- Run predefined rules/processes
- Analyze data and environment
- Determine best action
- Action:
- Respond to user via chat interface
- Running procedures via user's approval
- Log what has been done
- Ask user about incident resolution
- Reiterate if it's failed
- Conclude if it's resolved
- Trigger 3rd party integrations for resolution
- Inform stakeholders
- Trigger flags if there is any
- Add documentation
- Learning
- Record problem
- Record failure solution
- Record successful solutions into vector db
Created
May 5, 2025 12:39
-
-
Save erhangundogan/d24143cb57ef38fba0a288484ed742d8 to your computer and use it in GitHub Desktop.
Chatbot TODO
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment