camel-ai
[Feature Request] Implement Terminal Agent for Executing Terminal Commands
Required prerequisites
- [x] I have searched the Issue Tracker and Discussions that this hasn't already been reported. (+1 or comment there if it has.)
- [x] Consider asking first in a Discussion.
Motivation
An agent capable of executing terminal-related commands, such as inspecting local files or managing system processes. Adding a Terminal Agent will enable the system to perform a wide range of terminal operations dynamically.
Proposed Solution:
Develop a Terminal Agent capable of executing shell commands securely and returning the results. This agent should:
Handle common file system operations (e.g., list, read, and modify files).
Support command execution with appropriate permission control.
Return structured output to facilitate further processing.
Key Tasks:
Define the Terminal Agent with a secure command execution interface.
Implement basic file system and process management commands.
Add safeguards to prevent unauthorized or unsafe command execution.
Integrate the Terminal Agent into the workforce framework.
Test the Terminal Agent with various command scenarios.
Solution
No response
Alternatives
No response
Additional context
No response
Hi, @Aaron617 , I am interested in the issue, let me know if you need any help
Hey @Aaron617, could you help me understand the necessity of adding a new agent for terminal operations instead of directly providing the toolkit to the agent?
Maybe it's not necessary to build specific agent, but we can build a cookbook for this to showcase this
There are some specific feature beyond the toolkit itself. For example, the agent needs to inspect the file system at each timestep. However, I agree we could build a cookbook as the first step.
close this issue as terminal toolkit has been integrated: https://github.com/camel-ai/camel/blob/master/examples/toolkits/terminal_toolkit.py we can create another issue for cookbook writing