Forge AI: Intelligent Coding Agent
An AI coding agent that can inspect, reason about, and modify your codebase through an interactive command-line interface. Forge AI empowers developers by automating code inspection, proposing changes with user approval, and understanding complex project structures.
✨ Key Features
- Interactive CLI: Engage with the AI agent directly from your terminal.
- Intelligent Code Modification: Agent can inspect, reason about, and propose code changes.
- Human-in-the-Loop: All code changes are presented as diffs for user review and approval.
- Multi-LLM Support: Configurable to use various LLM providers (OpenAI, Google Gemini, Anthropic).
- Conversation Memory: Persists chat history across sessions using SQLite.
- Project Structure Awareness: Agent builds a dynamic understanding of your project layout.
- Data & Notebook Inspection: Specialized tools for summarizing datasets and reading Jupyter Notebooks.
- Code Execution Capabilities: Can execute the code and analyze the outputs or errors.
🚀 Installation & Usage
Prerequisites
- Python 3.8+
Installation
Install from PyPI:
pip install forge-ai
Configuration
Before using Forge AI, you need to initialize it with your preferred LLM provider and API key.
forge init --provider <provider_name> --api-key <YOUR_API_KEY> [--model <model_name>]
Example:
- OpenAI:
forge init --provider openai --api-key sk-xxxx --model gpt-4o-mini
- Google Gemini:
forge init --provider gemini --api-key your-gemini-api-key --model gemini-2.5-flash
- Anthropic:
forge init --provider anthropic --api-key your-anthropic-api-key --model claude-3-haiku-20240307
Forge AI will store your configuration in a .forge/config.json file in your current working directory.
Running the Agent
Simply run forge to start an interactive chat session with the AI agent:
forge
- Continue a Session: To pick up a previous conversation, use the
--thread-idoption (displayed when a new session starts).forge --thread-id <your_thread_id>
- Exit: Type
exitorquitto end the chat session. You'll be prompted to clear memory or delete all Forge data.
CLI Commands
forge init: Initialize Forge AI with your LLM provider and API key.forge init --provider openai --api-key YOUR_KEY
forge clear_memory: Clears all conversation history from the SQLite database.forge clear_memoryforge stop: Deletes the entire.forgedirectory, including configuration and all conversation memory. This is irreversible.forge stop
CHANGELOG
1.1.0 - 2025-09-16
- 🏗 Added notebook writing and code execution capabilites.
1.0.0 - 2025-09-14
- 🚀 Initial release
Metadata
Release files for forge-ai 1.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| forge_ai-1.1.0.tar.gz | 17.2 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| forge_ai-1.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 36.1 kB
Release files / forge_ai-1.1.0.tar.gz
| Download URL | forge_ai-1.1.0.tar.gz |
|---|---|
| Size | 17.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
25967b2ad1f9224c2618184901935e51322f25856588753b3857689d511574ac
|
|
BLAKE2b-256 checksum How to use checksums |
81183a06f14294eb36555d9e3cf05e7d3ebb8be63c47e068241c7dc87d59ac4e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 15, 2025.
Transparency logRelease files / forge_ai-1.1.0-py3-none-any.whl
| Download URL | forge_ai-1.1.0-py3-none-any.whl |
|---|---|
| Size | 18.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
9e5cd2844fedc1780102fd9d55c84191318161e4de50ad400c9d0aff572007aa
|
|
BLAKE2b-256 checksum How to use checksums |
f9f9feb62dc576194a78c8342c302837ecb848df549fc35ae146ab0ce3997609
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 15, 2025.
Transparency log