Testing Arwaky
Autonomous test engine with self-healing capabilities. Built for AI agents that write and maintain tests 24/7.
For AI Agents:
- Autonomous test generation and execution
- Self-healing test code (auto-fix failures)
- Coverage auditing and quality gating
- Synthetic test data generation
Choose Your Path
| I'm a... | Start Here | What I'll Do |
|---|---|---|
| AI Agent | SKILL.md | Autonomous testing, self-healing |
| Developer | Quick Start | Run tests, generate tests, audit coverage |
| Contributor | Contributing | Add capabilities, adapters, CLI commands |
Why Use Testing Arwaky
For Users
| Benefit | Description |
|---|---|
| Self-Healing | Automatically fixes broken tests |
| AST Analysis | Understands code structure to generate smart tests |
| Coverage Audit | Ensures test coverage meets thresholds |
| Data Generation | Creates edge-case test data automatically |
| CI-Ready | Exit codes and JSON output for pipelines |
The Cost of NOT Using Testing Arwaky
┌─────────────────────────────────────────────────────────────────┐
│ What you're losing right now: │
├─────────────────────────────────────────────────────────────────┤
│ ❌ Hours writing boilerplate tests manually │
│ ❌ Brittle tests that break on minor refactors │
│ ❌ Low coverage on edge cases │
│ ❌ Test suites that drift out of sync with code │
│ ❌ Manual test data creation for boundary conditions │
└─────────────────────────────────────────────────────────────────┘
Install
One-Liner (Linux/macOS)
curl -sSL https://raw.githubusercontent.com/rakaarwaky/testing-arwaky/main/install.sh | bash
One-Liner (Windows PowerShell)
iwr https://raw.githubusercontent.com/rakaarwaky/testing-arwaky/main/install.ps1 | iex
Manual Install
pip install testing-arwaky-mcp
Or with uv:
uv tool install testing-arwaky
Verify
agentic-test version
agentic-test --help
Quick Start
# Run tests with self-healing
agentic-test run tests/test_parser.py --heal --max-retries 3
# Analyze source code
agentic-test analyze src/my_module.py
# Audit coverage
agentic-test audit ./src --threshold 80
# Generate boilerplate tests
agentic-test generate src/my_module.py
# Generate synthetic test data
agentic-test generate-data emails --count 10
CLI Commands
Test Commands
| Command | Description |
|---|---|
agentic-test run <path> [--heal] [--max-retries N] |
Run tests with optional self-healing |
agentic-test analyze <file> |
AST analysis of source file |
agentic-test audit <dir> [--threshold N] |
Coverage audit with pass/fail gate |
agentic-test generate <file> |
Generate boilerplate tests from AST |
Data & Migration
| Command | Description |
|---|---|
agentic-test generate-data <type> [--count N] |
Generate synthetic test data |
agentic-test migrate <file> [--backup] |
Migrate unittest to pytest |
Performance & Workflow
| Command | Description |
|---|---|
agentic-test find-slow <dir> [--threshold N] |
Find slow tests |
agentic-test mock-generate "<signature>" |
Generate mock from function signature |
agentic-test workflow <name> <target> |
Run pre-defined workflows |
Utility
| Command | Description |
|---|---|
agentic-test version |
Show version |
agentic-test init <path> |
Initialize configuration file |
agentic-test governance <dir> |
Architectural governance scan |
Full list: agentic-test --help
Architecture
5-domain structure (same as lint-arwaky):
src/
├── agent/ # Lifecycle, orchestration, DI container
├── capabilities/ # Thinking logic — test execution, healing, analysis
├── infrastructure/ # Adapters — pytest runner, file system, transports
├── surfaces/ # Interfaces — CLI (Click), MCP (FastMCP)
└── taxonomy/ # Value objects, interfaces, models
Dependency Rules
surfaces → capabilities OK
surfaces → infrastructure OK
capabilities → infrastructure OK (uses interfaces from taxonomy)
infrastructure → taxonomy OK
agent → everything OK (wiring layer)
Self-Healing
Testing Arwaky can automatically fix common test failures:
| Error Type | Fix Strategy |
|---|---|
| ImportError | Add missing sys.path entries |
| AttributeError | Detect typos using Levenshtein distance |
| AssertionError | Patch expected values with actual values |
| TypeError | Add missing positional arguments |
| NameError | Insert common imports (os, sys, json, etc.) |
Example healing loop:
1. Run test → FAIL (ImportError)
2. Healer detects missing sys.path
3. Applies fix, retries → PASS
Configuration
Environment Variables
| Variable | Default | Description |
|---|---|---|
DC_SOCKET_PATH |
/tmp/dc.sock |
Unix socket path for ComanderArwaky |
Contributing
See CONTRIBUTING.md for details on:
- Adding new test capabilities
- Adding CLI commands
- Adding MCP tools
- Testing guidelines
License
MIT License. See LICENSE.
Release files for testing-arwaky 1.1.8
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| testing_arwaky-1.1.8.tar.gz | 42.6 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| testing_arwaky-1.1.8-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 91.6 kB
Release files / testing_arwaky-1.1.8.tar.gz
| Download URL | testing_arwaky-1.1.8.tar.gz |
|---|---|
| Size | 42.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
7650c3aca9950ab3e64f3e18e42008939dcb24537aed49b77168b04b4443ba21
|
|
BLAKE2b-256 checksum How to use checksums |
5a31be3375ed0e4ab1b2e9a7adc7ea58ae626c139af177a6e0103089dfc6c565
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
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 Jun 2, 2026.
Transparency logRelease files / testing_arwaky-1.1.8-py3-none-any.whl
| Download URL | testing_arwaky-1.1.8-py3-none-any.whl |
|---|---|
| Size | 49.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
885949c4e1185b42441bf90b587965057cd2b53159d4127b036bd2e1331b51c0
|
|
BLAKE2b-256 checksum How to use checksums |
51405a5b737c9b647186705e3d74d4396a525ce97b6d9df450e093c7a004d1a7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
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 Jun 2, 2026.
Transparency log