jiro-dreams-of-code
"I'll continue to climb, trying to reach the top, but no one knows where the top is." — Jiro Ono
A disciplined AI agent orchestration system that builds software using the shokunin philosophy: mastery through small, deliberate, repeated steps.
Philosophy
Named after Jiro Ono, the 93-year-old sushi master from "Jiro Dreams of Sushi," this project embodies:
- Shokunin (職人): The artisan spirit—dedication to mastering one craft through continuous refinement
- Kaizen: Continuous small improvements
- Discipline over speed: Correctness and reviewability trump velocity
Features
- Generate detailed specifications from natural language prompts
- Break specifications into granular, dependency-aware tasks
- Execute tasks with AI agents using strict TDD discipline
- Enforce commit hygiene through automated validation
- Track context usage and metrics for every operation
- Web UI for real-time monitoring and refinement
Installation
# Using uv (recommended)
uv pip install jiro-dreams-of-code
# Using pip
pip install jiro-dreams-of-code
Quick Start
# Initialize jiro in your project
jiro init
# Verify installation
jiro doctor
# Generate a spec from a prompt
jiro dream "build a user authentication system"
# Plan the implementation
jiro plan --spec auth-system.md # looks in .jiro-dreams-of-code/specs/auth-system.md by default
# Execute the tasks
jiro execute
# Monitor progress
jiro status
jiro web
Specs generated during jiro dream are saved in .jiro-dreams-of-code/specs/ (e.g., auth-system.md). jiro plan --spec accepts either the bare spec name for automatic resolution or a full path if you need to override the location.
Documentation
See TECHNICAL_SPEC.md for complete documentation.
Requirements
- Python 3.12+
- Git
- Anthropic API key (for Claude)
Anthropic Credentials
Commands that rely on Anthropic require a key that can be provided by either:
- Setting the
ANTHROPIC_API_KEYenvironment variable - Storing the key in the system keyring (
jiro doctorverifies both sources)
If both are configured, jiro doctor emits a warning so you can consolidate to one source. If neither source is available, Anthropic-dependent commands exit immediately with an error rather than prompting mid-task.
Development
# Clone the repository
git clone https://github.com/marshally/jiro-dreams-of-code.git
cd jiro-dreams-of-code
# Install with dev dependencies
uv pip install -e ".[dev,test]"
# Set up pre-commit hooks
pre-commit install
# Run tests
pytest
# Run linter
ruff check .
# Run type checker
mypy src/jiro
License
MIT
Metadata
Release files for jiro-dreams-of-code 0.1.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| jiro_dreams_of_code-0.1.1.tar.gz | 124.6 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| jiro_dreams_of_code-0.1.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 134.5 kB
Release files / jiro_dreams_of_code-0.1.1.tar.gz
| Download URL | jiro_dreams_of_code-0.1.1.tar.gz |
|---|---|
| Size | 124.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
b39ebd18756ca9ae129c35c9d4d084d3e46bc99c8e8dfdf3ddbf440c0ed11725
|
|
BLAKE2b-256 checksum How to use checksums |
a59e6f04d6cdc405e87f3f1367f66ac47708b795debc280a720883541d915d69
|
| 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 Nov 29, 2025.
Transparency logRelease files / jiro_dreams_of_code-0.1.1-py3-none-any.whl
| Download URL | jiro_dreams_of_code-0.1.1-py3-none-any.whl |
|---|---|
| Size | 9.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
2bf4034f2e9b6c6b6818093b400acf651a4f56d311694fa8788b6c6baa6c8daf
|
|
BLAKE2b-256 checksum How to use checksums |
6f3a50c6ea41821a7db539838632683063b2f06eef96af9c833e06a42ff45871
|
| 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 Nov 29, 2025.
Transparency log