KodeAgent
A lightweight Python engine for planning, tool-using, and code-executing AI agents.
- Source repo
- barun-saha/kodeagent
- Stars
- ★ 45
- Last updated
- 1mo ago
- License
- Apache-2.0
- Primary language
- Python
- FA score
- 49/100 · Major gaps
At a glance
- How it runs
- Works with
- Universal · cross-platformOpenAI API · Claude API
- Cost
- Free software; you pay for model usage
- Setup effort
- Low · running in minutes
- You'll need
- Typical use
- A Python-service developer building a web-research workflow can use ReActAgent with search_web and read_webpage for multi-step information gathering.
- Not a fit if
- Teams wanting a full platform with automated persistent long-term memory
- Users needing async tools or a batteries-included agent framework
- Anyone running LLM-generated code in production without review
- Source review
- 49/100 · Major gaps 2 safety controls not found
What does this agent do, and when should you use it?
KodeAgent is a frameworkless, minimal Python AI-agent engine intended to serve as the reasoning core inside a larger system rather than as a complete platform. It provides ReActAgent, CodeActAgent, and FunctionCallingAgent, using LiteLLM to work with models and providers such as Gemini, GPT, Claude, and Ollama. Agents stream updates asynchronously through agent.run(task), with the completed task result available through agent.task.result. CodeActAgent can execute LLM-generated Python on the local host or in an E2B sandbox, while CSVAnalysisAgent loads CSV data from a URL or path for automated analysis. Tasks are memoryless by default, but callers can use recurrent_mode for the preceding task's context or inject OpenAI-compatible chat history.
A developer creates a ReActAgent, CodeActAgent, or FunctionCallingAgent with a model_name, synchronous Python tools, and iteration settings, then consumes streamed responses with async for response in agent.run(task). ReActAgent can invoke tools such as search_web, read_webpage, and calculator directly. CodeActAgent creates a task plan, repeatedly asks the model for Thought and Python Code, executes tool calls, captures observations, and updates its Planner. It runs generated code either in the application's host environment or an e2b sandbox, with sandbox-created files downloadable to an existing local work_dir. Built-in tools cover file downloads, Markdown extraction, arXiv and Wikipedia search, image generation, and audio or YouTube transcription; developers can also pass their own synchronous Python functions as tools. Optional tracing_type='langfuse' or tracing_type='langsmith' sends traces to those services, while the default tracer is a no-op.
- A Python-service developer building a web-research workflow can use ReActAgent with search_web and read_webpage for multi-step information gathering.
- An engineering team that needs an LLM to write and run Python while combining tools can use CodeActAgent, choosing E2B for sandboxed execution.
- A developer using a small language model with native function calling can use FunctionCallingAgent for its documented retry logic and error detection.
- An analyst who needs to load a CSV from a URL or local path and explore it automatically can use CSVAnalysisAgent.
- An application team persisting conversations in its own database can resume context by passing OpenAI-compatible chat_history to agent.run.
How do you install or deploy this agent?
Run pip install -U kodeagent. The documented supported runtimes are Python 3.10, 3.11, 3.12, and 3.13. Set credentials for the selected provider, such as GOOGLE_API_KEY for Gemini, OPENAI_API_KEY for OpenAI, or ANTHROPIC_API_KEY for Anthropic; E2B execution also requires E2B_API_KEY. To run from source, use git clone https://github.com/barun-saha/kodeagent.git, create and activate a virtual environment, then run pip install -r requirements.txt.
How do you use this agent?
Import ReActAgent and the required tools, then create an instance such as ReActAgent(name='Web agent', model_name='gemini/gemini-2.5-flash-lite', tools=[search_web, read_webpage], max_iterations=5). Run async for response in agent.run('What are the festivals in Paris? How they differ from Kolkata?'): and print the final output with print_response(response, only_final=True). For generated-code execution, use CodeActAgent and explicitly set run_env, allowed allowed_imports, and, when needed, pip_packages. For E2B, an existing work_dir can be used as the local file-exchange directory.
What are this agent's strengths and limitations?
- It offers ReAct, CodeAct, and native function-calling agent modes through one Python-oriented API.
- Its LiteLLM integration lets applications switch among Gemini, GPT, Claude, and Ollama without changing core logic.
- CodeActAgent supports both local-host and E2B execution, including transfer of sandbox-generated files to a local workspace.
- OpenAI-compatible chat-history injection allows an external application to own persistent session state.
- The project states that it is actively evolving, so aspects may change and capabilities are limited.
- Agents are memoryless across tasks by default; there is no automated long-term memory management or persistence layer.
- CodeActAgent executes LLM-generated code, and the project explicitly advises reviewing and testing it in constrained, non-production environments.
- E2B requires an E2B API key; LangSmith is not installed by default and requires separate installation and credentials.
- Custom tools must be synchronous Python functions; asynchronous tools are unsupported.
How does this agent compare with similar options?
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| KodeAgent This agent | 49 · Major gaps | Library / SDKFree + model costs | ★ 45 | 1mo ago | Python | OpenAI API · Claude API |
| CaveAgent | 63 · Some gaps | Library / SDKFree + model costs | ★ 219 | 1mo ago | Python | OpenAI API · Claude API |
| AI Agents From Scratch | 55 · Major gaps | CLIFree | ★ 4.9k | 3mo ago | JavaScript | — |
| Langroid | 72 · Some gaps | Library / SDKFree + model costs | ★ 4.1k | 2d ago | Python | Claude Code · OpenAI API |
How does FollowAgents rate this agent?
Why each dimension lost points
The README identifies model providers, API keys, tracing services, tools, and host/E2B execution paths; tracing is off by default, so data flows are partly visible (2). However, CodeAct can run generated code, host mode uses the application environment directly, and the documented import whitelist and review checks are limited; the security document itself says the measures are not reliable. Least privilege, sensitive-data protection, and external-effects control are therefore partial (1 each). No pre-execution user confirmation or rollback mechanism is shown (0 each). Requirements specify version ranges and some compatibility notes, but provide no vulnerability-response evidence (1). The README acknowledges inspirations and the Apache license file is present; publisher identity is unverified and ownership/attribution detail remains limited (2).
README, usage, and security documentation broadly agree on execution modes and features, but examples and parameter status vary in detail, and strong security claims coexist with disclaimers, so consistency is partial (1). Requirements constrain some versions and describe a compatibility issue, while dependency services, optional packages, and external APIs limit availability (1). Documentation describes errors, timeouts, and iteration limits that can leave no result, and integration tests show skip behavior; failure handling is explained to a degree but not comprehensively evidenced (2).
The documentation distinguishes ReAct, CodeAct, and FunctionCallingAgent for smaller models, with task and model guidance (2). It describes default memory behavior, tracing limitations, code-execution risks, and several feature limits (2). Users can supply tools, but capability boundaries and triggering conditions largely depend on the model and task prompt, with no systematic suitability rules (1). Setup guidance covers Python 3.10+, virtual environments, host/E2B, environment variables, and optional tracing dependencies (2).
README and docs organize quick start, security, installation, and usage into clear topics (2). Installation and optional dependency steps are reasonably specific (2). Agent names are recognizable, but a deprecated parameter still appears in examples, leaving interface evolution uncertain (1). Examples cover multiple use patterns and model selection, and identify some poor-fit scenarios (2). The materials state that there is no persistence layer, the project is evolving, some model classes may not work, and host execution is risky (2). LICENSE and package metadata both identify Apache-2.0 (3). Version is dynamically loaded and README mentions v0.12.1, but no changelog is present in the supplied material (1). A bug tracker and release workflow are listed, but no ongoing maintainer responsibility or support commitment is clear (1).
Streaming responses, final results, plans, and token/cost metrics have concrete usage guidance, supporting integration and consumption (2). A small multi-paradigm engine, custom tools, and model flexibility provide clear value, while still relying on external models, tools, and services (2). The dependency set is comparatively focused and allows local or sandbox execution; CodeAct, model calls, and optional cloud services add runtime and operational costs, so the benefit case is supported but not fully quantified (2).
Many feature and configuration claims have matching descriptions across README, usage, and installation docs, and the security document outlines a processing flow (2). However, the supplied source sample omits core implementation, so security behavior and resource-limit claims are not independently corroborated here (1). The docs include limitations and warnings, but the security document also makes strong assurance-style claims alongside disclaimers, so fact and inference are not always clearly separated (1).
- Not found in source: confirmation before actingTurn on (or add) a confirmation step before it acts, and try it in a sandbox or test environment before real data.
- Not found in source: rollback or recovery pathBack up first, or work on a git branch or snapshot, so its changes can be undone.
- CodeAct host mode runs model-generated code on the application system. The docs recommend E2B, while also admitting that the safeguards are not reliable. Before production use, review execution permissions, secret visibility, and file/network boundaries.
- The actual security and availability of dependencies and external model, search, and tracing services cannot be established from this static packet; badges and documentation claims are not verification results.
Maintain this project? Add the FARS badge to your README
[](https://followagents.com/en/agents/kodeagent)<a href="https://followagents.com/en/agents/kodeagent"><img src="https://followagents.com/badge/kodeagent.svg" alt="FARS review"></a>Recent releases
Latest entries from the project's GitHub Releases. Notes are quoted from upstream as written, not rewritten or tested by FollowAgents.
-
Release notes excerpt
What's Changed * chore: Update libs and bump version by @barun-saha in https://github.com/barun-saha/kodeagent/pull/232 Full Changelog: https://github.com/barun-saha/kodeagent/compare/v0.14.3...v0.14.4View on GitHub ↗ -
Release notes excerpt
What's Changed * Upgrade Codecov action to v7 by @barun-saha in https://github.com/barun-saha/kodeagent/pull/228 * Refactor dotenv loading and logging configuration by @barun-saha in https://github.com/barun-saha/kodeagent/pull/227 * Add Ollama usage examples by @barun-saha in https://github.com/barun-saha/kodeagent/pull/229 * Upgrade packages and release new version by @barun-saha in https://github.com/barun-saha/kodeagent/pull/230 Full Changelog: https://github.com/barun-saha/kodeagent/compare/v0.14.2...v0.14.3View on GitHub ↗ -
Release notes excerpt
What's Changed * Patch libs, update model names and docs by @barun-saha in https://github.com/barun-saha/kodeagent/pull/225 Full Changelog: https://github.com/barun-saha/kodeagent/compare/v0.14.1...v0.14.2View on GitHub ↗
FAQ
Can it replace a full agent platform?
Is it tied to one model provider?
What are the safety boundaries for CodeActAgent?
Does the repository document operating costs?
Thanks — we use this to improve the page.