Skaro
The open-source spec-driven workspace
for software development with AI
Website · Documentation · PyPI · Telegram · Discord
What is Skaro?
Skaro is positioned as a tool where the developer remains the architect and the AI acts as the executor: the platform supports architecture reviews, ADRs, DevPlans, step-by-step task execution, Git integration, model usage analytics, and stack-specific instruction sets for different technologies. That makes Skaro a practical orchestration layer for AI-assisted software development, especially in projects where reproducibility, consistency, and quality control matter.
Features
- Project artifacts live next to the code — constitution, architecture, ADRs, development plans, and task specs are stored in .skaro/ inside the repository, so project context stays versioned with the codebase.
- Fast onboarding for existing repositories — Skaro can analyze an existing codebase and generate initial artifacts such as constitution, architecture, and an inventory of already implemented functionality.
- Engineering rules are explicit — project conventions, stack constraints, and architectural decisions are captured as real artifacts, so AI works within defined boundaries instead of relying on ad hoc prompting.
- Ideas become executable plans — Skaro turns features and changes into milestones and tasks, making the implementation path explicit instead of leaving it scattered across chats.
- Tasks move through a fixed workflow — each task follows clarify → plan → implement → tests, helping teams avoid jumping straight into code generation without alignment and structure.
- AI works with repository-aware context — Skaro selects relevant files and combines them with project structure, so the model gets focused context for the current step instead of the entire codebase at once.
- Completion is verified, not assumed — tasks can include structural checks, test commands, and recorded validation results, so “done” means reviewed and verified.
- Project-wide review is built in — beyond task-level execution, Skaro can validate project artifacts, task states, and verification steps across the whole repository.
- Git stays part of the workflow — diffs, staging, commits, and branch operations are integrated into the process, keeping implementation flow tied to the actual repository state.
- AI behavior is configurable for the stack — models, providers, skills, and stack-specific instruction sets make it possible to adapt AI execution to the technology and engineering style of the project.
- LLM usage is visible — usage statistics show token consumption by role, phase, task, and model, making AI cost and workflow patterns easier to understand.
Install
Python 3.11+ required. Everything included: CLI, web dashboard, LLM adapters, templates.
Linux / macOS:
curl -fsSL https://raw.githubusercontent.com/skarodev/skaro/main/install.sh | sh
Windows (PowerShell):
irm https://raw.githubusercontent.com/skarodev/skaro/main/install.ps1 | iex
Alternative (if you have pipx or uv):
pipx install skaro
# or
uv tool install skaro
Quick start
cd my-project
skaro init
skaro ui
skaro init creates a .skaro/ directory with constitution, architecture template, and config.
skaro ui starts the web dashboard at http://localhost:4700. LLM provider is configured from the UI.
Update
Check for a new version:
skaro update
Use --force to bypass the 24-hour cache:
skaro update --force
Upgrade — install script (venv):
| OS | Command |
|---|---|
| Windows | & "$env:USERPROFILE\.skaro\venv\Scripts\pip.exe" install --upgrade skaro |
| macOS / Linux | ~/.skaro/venv/bin/pip install --upgrade skaro |
Or simply re-run the install script — it detects the existing venv and upgrades in place.
Upgrade — pipx:
pipx upgrade skaro
Verify after upgrade:
skaro --version
From source (development)
git clone https://github.com/skarodev/skaro.git
cd skaro
python3 -m venv .venv && source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install -e ".[dev]"
Frontend (requires Node.js 18+):
cd frontend
npm install
npm run build
Run tests:
pytest
License
AGPL-3.0 — see LICENSE.
From Russia with love ❤️
1 maintainer
Metadata
1 maintainer
Release files for skaro 2.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| skaro-2.0.1.tar.gz | 1.5 MB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| skaro-2.0.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 2.8 MB
Release files / skaro-2.0.1.tar.gz
| Download URL | skaro-2.0.1.tar.gz |
|---|---|
| Size | 1.5 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
56e17f7bc04f5cfaaac8115b6556349e70f0ee8376c1ae2d97732ca32b933d40
|
|
BLAKE2b-256 checksum How to use checksums |
38bbfc86926a7da9133a60fc9b83e97c80f61a0243b6002d603e30ae974297ca
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Release files / skaro-2.0.1-py3-none-any.whl
| Download URL | skaro-2.0.1-py3-none-any.whl |
|---|---|
| Size | 1.2 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
ba571dfdd4fd8656610008b6672baea6e786d5a4bfc550b4c9ff77f84f2356f4
|
|
BLAKE2b-256 checksum How to use checksums |
d8bc2716cefad18653baa611ce0babef3d670b4b75e6fdda11a27da30a3b6990
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Security for skaro 2.0.1
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.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
