🤖 KodeAgent

KodeAgent: A minimalistic approach to building AI agents.

✅ Why KodeAgent?

Here are some reasons why you should use KodeAgent:

Framework-less: Unlike some heavy agentic frameworks, KodeAgent stays lightweight, making it easy to integrate and extend.
Learn-first design: Helps developers understand agent-building from scratch.
Multimodal: Supports both text and images in the inputs.

Written in about 2000 lines (excluding the prompts), KodeAgent comes with built-in ReAct and CodeAct agents. Or you can create your own agent by subclassing Agent.

A key motivation beyond KodeAgent is also to teach building agentic frameworks from scratch. KodeAgent introduces a few primitives and code flows that should help you to get an idea about how such frameworks typically work.

✋ Why Not?

Also, here are a few reasons why you shouldn't use KodeAgent:

KodeAgent is actively evolving, meaning some aspects may change.
The first priority is simplicity; optimization is secondary.
You already use some of the well-known frameworks or want to use them.

👨‍💻 Usage

Clone the KodeAgent GitHub repository locally:

git clone https://github.com/barun-saha/kodeagent.git

Next, create a virtual environment if you do not have one already and activate it:

python -m venv venv
source venv/bin/activate
# venv\Scripts\activate.bat  # Windows

KodeAgent has only a few direct dependencies. Install them as follows:

pip install -r requirements.txt

Now, in your application code, create a ReAct agent like this:

from kodeagent import ReActAgent


agent = ReActAgent(
    name='Maths agent',
    model_name='gemini/gemini-2.0-flash-lite',
    tools=[calculator],
    max_iterations=3,
)

Or if you want to use CodeAct agent:

from kodeagent import CodeActAgent

agent = CodeActAgent(
    name='Web agent',
    model_name='gemini/gemini-2.0-flash-lite',
    tools=[search_web, extract_file_contents_as_markdown],
    run_env='e2b',
    max_iterations=3,
    allowed_imports=['re', 'requests', 'duckduckgo_search', 'markitdown'],
    pip_packages='ddgs~=9.5.2;"markitdown[all]";',
)

Now let your agent solve the tasks like this:

for task in [
    'What is 10 + 15, raised to 2, expressed in words?',
]:
    print(f'User: {task}')

    async for response in agent.run(task):
        print_response(response)

That's it! Your agent should start solving the task and keep streaming the updates. For more examples, including how to provide files as inputs, see the kodeagent.py module.

KodeAgent uses LiteLLM, enabling it to work with any capable LLM. Currently, KodeAgent has been tested with Gemini 2.0 Flash Lite. For advanced tasks, you can try Gemini 2.5 Pro.

LLM model names, parameters, and keys should be set as per LiteLLM documentation. For example, add GEMINI_API_KEY to the .env to use Gemini API.

Code Execution

CodeActAgent executes LLM-generated code to leverage the tools. KodeAgent currently supports two different code run environments:

host: The Python code will be run on the system where you created this agent. In other words, where the application is running.
e2b: The Python code will be run on an E2B sandbox. You will need an E2B API key and add to your .env file.

With host as the code running environment, no special steps are required, since it uses the current Python installation. However, with e2b, code (and tools) are copied to a different environment and execute. Therefore, some additional set up may be required.

For example, the Python modules that are allowed to be used in code should be explicitly specified using allowed_imports. In addition, any additional Python package that may need to be installed should be specified as a comma-separated list via pip_packages.

KodeAgent is very much experimental. Capabilities are limited. Use with caution.

Sequence Diagram for CodeAct Agent (via CodeRabbit)

sequenceDiagram
  autonumber
  actor User
  participant Agent
  participant Planner
  participant LLM as LLM/Prompts
  participant Tools

  User->>Agent: run(task)
  Agent->>Planner: create_plan(task)
  Planner->>LLM: request AgentPlan JSON (agent_plan.txt)
  LLM-->>Planner: AgentPlan JSON
  Planner-->>Agent: planner.plan set

  loop For each step
    Agent->>Planner: get_formatted_plan()
    Agent->>LLM: codeact prompt + {plan, history}
    LLM-->>Agent: Thought + Code
    Agent->>Tools: execute tool call(s)
    Tools-->>Agent: Observation
    Agent->>Planner: update_plan(thought, observation, task_id)
  end

  Agent-->>User: Final Answer / Failure (per codeact spec)

Run Tests

To run unit tests, use:

python -m pytest .\tests\unit -v --cov --cov-report=html

For integration tests involving calls to APIs, use:

python -m pytest .\tests\integration -v --cov --cov-report=html

Gemini and E2B API keys should be set in the .env file for the tests to work.

Note: Some of the unit tests still make calls to the LLM. This would be fixed in the future.

🗺️ Roadmap & Contributions

To be updated.

🙏 Acknowledgement

KodeAgent heavily borrows code and ideas from different places, such as: