MARX - Multi-Agentic Review eXperience
Marx spins up multiple AI agents in Docker containers, each reviewing your PR independently. It merges their findings, removes duplicates, and gives you a pending GitHub review you can edit before submitting.
Screenshots
Features
- Parallel multi-agent reviews with automatic deduplication
- YOLO-mode agents run in Docker with review tools and PR context
- Create a pending GitHub review from merged issues (edit before submit)
- Structured review outputs with a merged summary
- Interactive PR selection (excludes your own PRs and PRs assigned to you)
- Works with local CLI configs or API keys (config dirs are mounted into containers)
- JSON output for automation via
--json-output
Prerequisites
git- gh (authenticated)
docker
Install
uv tool install marx-ai
Need Nix or a source install? See Installation.
Configure
Create ~/.marx with your GitHub token and any agent keys you want to use.
The GitHub token must be a classic PAT with the repo scope
(details):
cat > ~/.marx <<'MARX'
GITHUB_TOKEN=ghp_your_token_here
ANTHROPIC_API_KEY=your_claude_key
OPENAI_API_KEY=your_openai_key
GEMINI_API_KEY=your_gemini_key
MARX
If you already use the agent CLIs locally, Marx copies ~/.claude, ~/.codex, and ~/.gemini
into the containers so those configs work there too. API keys are optional.
See Configuration for details.
Use
# Interactive PR selection
marx
# Review a specific PR with all agents
marx --pr 123 --agents claude,codex,gemini
# Machine-readable output
marx --pr 123 --json-output
Docs
Metadata
Release files for marx-ai 0.4.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| marx_ai-0.4.1.tar.gz | 33.9 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| marx_ai-0.4.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 64.5 kB
Release files / marx_ai-0.4.1.tar.gz
| Download URL | marx_ai-0.4.1.tar.gz |
|---|---|
| Size | 33.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
97a6416567a8731844eb62a53680832e9a44f1b603761badd9b7268d17b9604f
|
|
BLAKE2b-256 checksum How to use checksums |
6ad84078d306819b7a3e9bfcc5bfb1e5c1d10f90884953a883176d00b94d79b4
|
| 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 Feb 26, 2026.
Transparency logRelease files / marx_ai-0.4.1-py3-none-any.whl
| Download URL | marx_ai-0.4.1-py3-none-any.whl |
|---|---|
| Size | 30.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
920603b6aee7183c6fa3803dd4dab467cc8da354e428d337174d2e074e740abf
|
|
BLAKE2b-256 checksum How to use checksums |
1228ad81b07f9892b5b3f24288d91d42b4b4e68079caf3bc5fc9dbdafc1d3cdf
|
| 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 Feb 26, 2026.
Transparency log