Skip to content

Commit

Permalink
feat: add basic agent section
Browse files Browse the repository at this point in the history
  • Loading branch information
sgomez committed Oct 13, 2024
1 parent a6caeae commit c371fc2
Show file tree
Hide file tree
Showing 3 changed files with 41 additions and 9 deletions.
33 changes: 33 additions & 0 deletions docs/agents/basic.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,33 @@
# Understanding Agents

In the context of AI and chatbots, agents are "intelligent" entities designed to perform tasks autonomously or semi-autonomously. They leverage tools and function calling to enhance their capabilities, allowing them to interact with users and perform complex operations efficiently.

## What are Agents?

Agents can be thought of as specialized programs that respond to user inputs or environmental stimuli. They possess certain characteristics:

- **Autonomy**: Agents can operate without direct human intervention, making decisions based on predefined logic or learned behaviors.
- **Reactivity**: They can respond to changes in their environment, allowing them to adapt their behavior based on real-time information.
- **Proactivity**: Some agents can take initiative, anticipating user needs or system requirements and acting accordingly.

## Tools and Function Calling

Agents utilize tools and function calling mechanisms to perform specific tasks. These tools can be thought of as functions or APIs that agents can invoke to achieve their objectives. Here are some key concepts:

### Tools

- **Definition**: Tools are pre-defined functions or APIs that agents can access to carry out specific operations. They provide agents with the ability to interact with external systems, databases, or services.
- **Examples**: Common tools include APIs for fetching data, processing transactions, sending notifications, and more. In chatbot scenarios, tools might also include integrations with platforms like Google Calendar for scheduling or payment gateways for processing transactions.

### Function Calling

- **Definition**: Function calling refers to the mechanism by which agents invoke tools to perform tasks. This can involve sending requests, handling responses, and processing the results.
- **Process**:
1. **Identify**: The agent identifies a need or task that requires external assistance.
2. **Invoke**: The agent calls the appropriate function/tool to fulfill the request.
3. **Receive Response**: The agent waits for a response from the tool, which may include data or confirmation of the completed task.
4. **Act**: Based on the response, the agent takes further actions, such as updating a user or executing additional tasks.

## Conclusion

Agents, equipped with tools and the ability to call functions, provide a powerful means of enhancing user interaction and automating processes. By leveraging these capabilities, agents can deliver more responsive and intelligent services, ultimately improving the overall user experience.
4 changes: 0 additions & 4 deletions docs/agents/index.md

This file was deleted.

13 changes: 8 additions & 5 deletions mkdocs.yml
Original file line number Diff line number Diff line change
Expand Up @@ -119,7 +119,7 @@ markdown_extensions:

# Page tree
nav:
- Index: index.md
- Introduction: index.md
- Our first bot:
- Create a bot: bot/create-bot.md
- Cloning the template: bot/cloning.md
Expand All @@ -129,9 +129,12 @@ nav:
- Create the model register: chatbot/register.md
- Running a basic chatbot: chatbot/basic.md
- Adding Memory : chatbot/memory.md
- Data Augmentation:
- Data Augmentation: rag/index.md
- Enhanced Context with Data:
- Basic Data Augmentation: rag/basic.md
- Retrieval-Augmented Generation: rag/rag.md
- Dynamic Data Augmentation with Tools: rag/tools.md
- Agents:
- Agents: agents/index.md
- Agents: agents/basic.md
- Full code: full-code.md



0 comments on commit c371fc2

Please sign in to comment.