Thanks to visit codestin.com
Credit goes to github.com

Skip to content
/ chant Public

Task automation system with AI-powered execution. Declarative task definitions, automatic dependency handling, and LLM-driven implementation.

License

Notifications You must be signed in to change notification settings

lex00/chant

Chant

Chant

Idempotent Intention

CI Release codecov License: Apache-2.0 Latest Release API Docs

Chant is a spec execution platform for AI-assisted development. Specs are markdown files that agents execute.

Installation

See the Installation Guide for detailed platform-specific instructions.

Documentation

Full documentation is available at lex00.github.io/chant

To build and preview docs locally:

just docs-serve

Quick Start

1. Initialize your project:

chant init

2. Create a spec:

chant add "Add user authentication"

3. Execute the spec:

chant work 001

For detailed setup instructions, see the Quick Start Guide.

Core Concepts

  • Specs - Markdown files with YAML frontmatter describing work to be done
  • Execution - Agents implement specs following acceptance criteria
  • Verification - Continuous checking that work meets intent
  • Drift Detection - Know when reality diverges from specs

Key Features

  • Parallel execution - Run multiple specs concurrently with isolated worktrees
  • Chain execution - Process specs sequentially (chant work --chain) or chain through specific IDs
  • Approval workflow - Gate spec execution with human approval, auto-detect agent-assisted work
  • Branch mode - Execute specs in feature branches with automatic merge

Examples

See the examples/ folder for real-world workflows.

Development

# Build the binary
just build

# Run tests
just test

# Build and serve docs
just docs-serve

License

Apache-2.0. See LICENSE for details.

About

Task automation system with AI-powered execution. Declarative task definitions, automatic dependency handling, and LLM-driven implementation.

Topics

Resources

License

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Packages

 
 
 

Contributors 2

  •  
  •  

Languages