Description
Chisel AI is designed for solo founders and small teams who are building projects with AI agents. As codebases grow rapidly, it becomes challenging to keep track of all decisions and architectural intents. Chisel AI addresses this by providing a structured way to capture and preserve reasoning through specs and docs. The specs act as artifacts that capture intent, constraints, and implementation notes, making them accessible to both humans and AI agents. The docs serve as a permanent knowledge base, structured for easy retrieval and optimized for AI-ready schemas.
Chisel AI is currently in public alpha, inviting users to sign up for release notes. It offers a lifecycle-driven spec tracker, allowing users to manage features and decisions as local Markdown files. The platform is built in Rust, ensuring fast response times and efficient navigation through its terminal UI. Chisel AI supports a mode optimized for LLM context windows, making it compatible with various shell-capable agents.
The platform is open source, licensed under FSL-1.1, and converts to Apache 2.0 after two years. It is designed to be extended, allowing users to integrate it into scripts or AI agents using Machine Mode YAML output. Chisel AI is ideal for those who need to maintain decision memory and structured context, ensuring that projects scale without losing architectural intent.
Chisel AI's Core Features
Text-first tools for shaping information
Structured specs and docs for LLM use
Preserves project reasoning and decision memory
Lifecycle-driven spec tracker
Markdown-first knowledge base
Local SQLite index for full-text search
Interactive terminal UI for developers
Open source with FSL-1.1 license
How to use Chisel AI?
Configure: Set up Chisel AI with your project
Use: Capture specs and docs for AI agents
Retrieve: Utilize full-text search for knowledge
Extend: Integrate with scripts using YAML output
Chisel AI's Use Cases
- Project management
- AI agent integration
- Knowledge preservation
- Decision tracking
- Fast retrieval
