Muni CLI

The command-line interface for Muni. Run scientific tools on managed GPUs, inspect and download job outputs, and build canvases of tables, charts, structures, molecules, and code from a terminal, a script, or an AI agent.

Install

pipx install muni

Or pip install muni into an existing environment. Requires Python 3.11+. muni --version shows what you have; muni update --check compares it with the latest release.

muni login            # opens your browser; click Authorize
muni login --device   # no browser: prints a code to enter at app.muni.bio/device
muni whoami

For CI and agents, create a personal access token once and export it instead:

muni token create --name ci --ttl 90d
export MUNI_TOKEN=...   # the token printed above

More in muni help auth: scoped and space-pinned tokens, revocation, and profiles.

Run a tool

muni tools -q rowan                              # find a tool
muni tool rowan_solubility --inputs              # its parameters
muni run rowan_solubility smiles=CCO --dry-run   # validate and estimate cost; submits nothing
muni run rowan_solubility smiles=CCO --follow    # submit and stream status until done

Parameters are KEY=VALUE pairs parsed as JSON when possible (temperatures='[298.15, 310]'); @file reads a value from a file and --params-file params.json loads them all. muni run ipsae --batch jobs.jsonl submits one job per line.

Read results

muni job list --tool rowan_solubility -n 5
muni job status JOB_ID
muni wait JOB_ID                           # block until the job finishes
muni job summary JOB_ID                    # columns, row count, file count
muni job query JOB_ID --columns rank,sequence --limit 10
muni job files JOB_ID
muni job download JOB_ID results/summary.csv --output summary.csv
muni job download-all JOB_ID --glob "*.pdb" --output-dir ./structures/

Work on a canvas

A space holds pages, and a page holds nodes: tables, charts, structures, compounds, text, images, PDFs, JSON, code, tool jobs, and pipelines.

muni space list
muni page create "Results" --space SPACE_ID --json
muni run rowan_solubility smiles=CCO --page PAGE_ID --follow   # a job node plus result nodes
muni job materialize JOB_ID --page PAGE_ID                      # put an earlier job's results on a page
muni node create table "Screen" --file ./results.csv --page PAGE_ID
muni node list --page PAGE_ID
muni node type NODE_ID                     # its kind and the commands for it
muni node table query NODE_ID "select * from rows limit 5"

muni run without --page or --space runs detached: results stay in job storage until you materialize them. More in muni help targeting and muni help nodes.

Guides

The CLI ships its own documentation. muni help lists the topics:

Topic	Covers
`muni help agents`	the contract for scripts and AI agents: output, exit codes, IDs, spending
`muni help auth`	browser, device, and token sign-in; profiles
`muni help targeting`	spaces, pages, the directory binding, where results land
`muni help jobs`	tools, parameters, costs, waiting, outputs
`muni help nodes`	creating, editing, running, and connecting canvas nodes
`muni help pipelines`	authoring, releasing, and running pipelines
`muni help campaigns`	lab campaigns with Adaptyv
`muni help autopipe`	reading AutoPipe runs and results
`muni help exit-codes`	exit codes and error codes
`muni help environment`	environment variables
`muni help reference`	every command and flag (`--json` for a machine-readable tree)

muni COMMAND --help explains any single command.

For AI agents

Every command accepts --json and returns {"schema": 1, "data": ...} or an error object with a stable code, and exit codes are documented. Point your agent at:

muni help agents

Python SDK

The package also installs a Python client:

from muni import MuniClient

client = MuniClient()
job = client.jobs.submit("rowan_solubility", {"smiles": "CCO"})
result = client.jobs.wait(job.job_id, timeout=600)

See docs/sdk.md for the client, its namespaces, and errors.

Contributing

See AGENTS.md for the repository layout, conventions, tests, and releases, and CHANGELOG.md for what changed.

License

MIT

Release files for muni 0.2.15

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for muni 0.2.15
File	Size	Uploaded
muni-0.2.15.tar.gz	501.0 kB	Sep 23, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for muni 0.2.15
File	Interpreter	ABI	Platform	Reset
muni-0.2.15-py3-none-any.whl 388.0 kB Sep 23, 2026	Python 3	none	any	Details

Total release size: 889.0 kB

Release files / muni-0.2.15.tar.gz

Download URL	muni-0.2.15.tar.gz
Size	501.0 kB
Tags	Source
SHA-256 checksum How to use checksums	`1ec9f2049bb2818106304c17442bc22bce067d10b50174376abd61eb0c71f4cd`
BLAKE2b-256 checksum How to use checksums	`c3957ad9aff7e46d082b8425077ad3dc74a8bcf84381d6f9e64a32f1d8449ad8`
Upload date	Sep 23, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

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.

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/bioArena/muni-cli
Publishing commit: github.com/bioArena/muni-cli/tree/a92c94a54e33094fb55897faced0ea4d194cfa83

Workflow

Publishing configuration: github.com/bioArena/muni-cli/blob/a92c94a54e33094fb55897faced0ea4d194cfa83/.github/workflows/publish.yml
Publishing logs: github.com/bioArena/muni-cli/actions/runs/35831591103/attempts/1

Artifact

Subject: muni-0.2.15.tar.gz
SHA-256 checksum: 1ec9f2049bb2818106304c17442bc22bce067d10b50174376abd61eb0c71f4cd Verifying attestations

Release files / muni-0.2.15-py3-none-any.whl

Download URL	muni-0.2.15-py3-none-any.whl
Size	388.0 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`a77a2fde3947c660dde5fc50e184569972f45bba8e4e509d4133c36045bbd093`
BLAKE2b-256 checksum How to use checksums	`db9b5d104c30c0a51cde87047259ffd8c7b0bbdc21fd4ebf4a802263a2765acd`
Upload date	Sep 23, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Provenance

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/bioArena/muni-cli
Publishing commit: github.com/bioArena/muni-cli/tree/a92c94a54e33094fb55897faced0ea4d194cfa83

Workflow

Publishing configuration: github.com/bioArena/muni-cli/blob/a92c94a54e33094fb55897faced0ea4d194cfa83/.github/workflows/publish.yml
Publishing logs: github.com/bioArena/muni-cli/actions/runs/35831591103/attempts/1

Artifact

Subject: muni-0.2.15-py3-none-any.whl
SHA-256 checksum: a77a2fde3947c660dde5fc50e184569972f45bba8e4e509d4133c36045bbd093 Verifying attestations