Agent skill

Metaxy

by anam-org in anam-org/metaxy

This skill should be used when the user asks to "define a feature", "create a BaseFeature class", "track feature versions", "set up metadata store", "field-level lineage", "FieldSpec", "FeatureDep"…

Apache-2.0Auto-check passedDevelopment

Install Metaxy

skills CLI
$ npx skills add anam-org/metaxy --skill metaxy -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install anam-org/metaxy metaxy --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/anam-org/metaxy.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude-plugin/skills/metaxy .claude/skills/metaxy && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
metaxy
GitHub stars
124
Token cost
~1.5k tokens
SKILL.md length
410 words
Files
6
Skills in repo
8
Repo updated
First seen
Licence
Apache-2.0

At a glance

This skill should be used when the user asks to "define a feature", "create a BaseFeature class", "track feature versions", "set up metadata store", "field-level lineage", "FieldSpec", "FeatureDep"…

  • Asks to define a feature
  • SKILL.md covers Core Concepts, CLI, Multi-Environment & Feature… and Testing, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • Create a BaseFeature class

What it does

Metaxy is an agent skill from anam-org/metaxy. This skill should be used when the user asks to "define a feature", "create a BaseFeature class", "track feature versions", "set up metadata store", "field-level lineage", "FieldSpec", "FeatureDep", "run metaxy CLI", "metaxy migrations", "metaxy lock", "lock features", "external features", "multi-environment", "monorepo features", "enable Map datatype", "enablemapdatatype", or needs guidance on metaxy feature definitions, versioning, metadata stores, CLI commands, testing patterns, feature locking, Map datatype…

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 6 other files (for example `examples/cli.md`, `examples/configuration.md` and `examples/feature-definitions.md`).

It sits in Development, covering Test strategy and Monorepo tooling. It works with Python. The repository describes itself as: Pluggable metadata management framework for versioned incremental multimodal data/ML pipelines. The licence is Apache-2.0.

When your agent uses it

  • Asks to define a feature
  • Create a BaseFeature class
  • Track feature versions
  • Set up metadata store

Example prompts

  • “define a feature”
  • “create a BaseFeature class”
  • “track feature versions”
  • “/metaxy”

Requirements

  • Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 8337842. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are python and bash).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • docs.metaxy.io

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Metaxy loads about 1.5k tokens when it runs. Until then it costs about 144 tokens; SKILL.md has 410 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~144
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from anam-org/metaxy at commit 8337842, republished under its Apache-2.0 licence (© anam-org). 410 words, ~1,534 tokens.

Download SKILL.mdSave it as .claude/skills/metaxy/SKILL.md (or your agent's skills folder). This skill also uses 5 other files; get the full folder from GitHub.
name
metaxy
description
This skill should be used when the user asks to "define a feature", "create a BaseFeature class", "track feature versions", "set up metadata store", "field-level lineage", "FieldSpec", "FeatureDep", "run metaxy CLI", "metaxy migrations", "metaxy lock", "lock features", "external features", "multi-environment", "monorepo features", "enable Map datatype", "enable_map_datatype", or needs guidance on metaxy feature definitions, versioning, metadata stores, CLI commands, testing patterns, feature locking, Map datatype configuration, or multi-environment configuration.

Metaxy

Metaxy is a metadata layer for multimodal Data and ML pipelines that manages and tracks feature versions, dependencies, and data lineage across complex computational graphs.

Core Concepts

Feature Definitions

To define a feature, create a class inheriting from mx.BaseFeature with a FeatureSpec metaclass argument:

python
import metaxy as mx


class Video(
    mx.BaseFeature,
    spec=mx.FeatureSpec(
        key="media/video",
        id_columns=["video_id"],
        fields=["audio", "frames"],  # Logical fields: describe data contents for versioning
    ),
):
    # Metadata columns: stored in the metadata store, tracked by metaxy
    video_id: str
    path: str
    duration: float
    height: int
    width: int

Important distinction: fields in FeatureSpec are logical field specs that describe the data contents for versioning and lineage tracking. Class attributes are metadata columns stored in the metadata store. They serve different purposes and should not overlap.

To add dependencies between features, use the deps parameter with FeatureDep. To specify field-level lineage (for partial data dependencies), use FieldSpec with FieldDep or FieldsMapping.

Data Versioning

Metaxy automatically tracks sample versions and propagates changes through the dependency graph. To trigger recomputation when code changes, set code_version on FieldSpec:

python
fields = [
    mx.FieldSpec(key="embedding", code_version="2"),  # Bump to invalidate downstream
]
Metadata Stores

To configure a metadata store, create a metaxy.toml file or use programmatic configuration:

python
config = mx.MetaxyConfig(
    stores={"dev": mx.StoreConfig(
        type="metaxy.ext.polars.handlers.delta.DeltaMetadataStore",
        config={"root_path": "/tmp/metaxy"},
    )}
)
with config.use() as cfg:
    store = cfg.get_store("dev")

Supported backends: DuckDB, ClickHouse, BigQuery, LanceDB, Delta Lake.

Feature Graph

To visualize and manage the feature dependency graph, use the CLI:

bash
mx graph render            # Terminal visualization
mx push --store dev        # Push graph to store

CLI

Metaxy provides a CLI (metaxy or mx alias) for managing features, metadata, and migrations:

bash
mx list features --verbose     # List features with dependencies
mx graph render                # Visualize feature graph
mx metadata status --all-features  # Check metadata freshness (expensive!)
mx mcp                         # Start MCP server for AI assistants

Multi-Environment & Feature Locking

Cross-project dependencies typically resolve automatically via Python packages — if project B depends on project A as a Python dependency, its features are discovered at import time. Feature locking (mx lock) is needed for multi-environment setups where projects cannot be installed into each other (e.g., separate deployment environments). In that case, use mx push to publish definitions to a shared store, and mx lock to fetch them into a metaxy.lock file. Set locked = true (or METAXY_LOCKED=1) in production to enforce version consistency.

For configuration patterns and CLI usage, see examples/configuration.md and examples/cli.md.

Show full SKILL.md (134 more words)Show less

Testing

To test features in isolation, use context managers to avoid polluting the global registry:

python
import pytest
import metaxy as mx


@pytest.fixture
def metaxy_env(tmp_path):
    with mx.FeatureGraph().use():
        with mx.MetaxyConfig(
            stores={"test": mx.StoreConfig(
                type="metaxy.ext.polars.handlers.delta.DeltaMetadataStore",
                config={"root_path": str(tmp_path / "delta_test")},
            )}
        ).use() as config:
            yield config
Map Datatype (Experimental)

To enable native Arrow Map column support (recommended for stores that support it), set enable_map_datatype = true in metaxy.toml. Requires the polars-map package. See https://docs.metaxy.io/stable/guide/concepts/metadata-stores/#map-datatype

When working with Map columns, use metaxy.utils.collect_to_polars, metaxy.utils.collect_to_arrow, or metaxy.utils.switch_implementation_to_polars to materialize or convert frames. These utilities preserve Map column types that would otherwise be lost with standard Narwhals backend conversions. See https://docs.metaxy.io/stable/guide/concepts/metadata-stores/#map-datatype

Examples

For complete code examples, see:

  • examples/feature-definitions.md - Feature classes with dependencies and field-level deps
  • examples/configuration.md - TOML and programmatic configuration
  • examples/metadata-stores.md - Store operations
  • examples/testing.md - Test isolation patterns
  • examples/cli.md - CLI command reference

Documentation

For comprehensive documentation: https://docs.metaxy.io/stable/

Key pages:

© anam-org, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 5 other files in .claude-plugin/skills/metaxy of anam-org/metaxy.

  • SKILL.md
  • examples/cli.md
  • examples/configuration.md
  • examples/feature-definitions.md
  • examples/metadata-stores.md
  • examples/testing.md

Open the folder on GitHubat commit 8337842

Compare with similar skills

Metaxy next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Metaxy compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Metaxy this skillanam-org/metaxy124—~1.5kAutomated safety check: PassApache-2.0
Deply ConfigVashkatsi/deply173—~945Automated safety check: PassMIT
Tbdjlevy/strif131—~3.5kAutomated safety check: PassMIT
Light Project StructureLight0305/Light-skills640—~3kAutomated safety check: NotesMIT
Claude Settings Auditgetsentry/skills1k4 repos~3kAutomated safety check: PassApache-2.0
Rocky Devrocky-data/rocky304—~2.1kAutomated safety check: PassApache-2.0

Similar skills

  • Deply Config

    Vashkatsi/deply

    A skill your agent uses when creating, auditing, validating, or integrating Deply v1.2.0 architecture configuration files for Python projects, including deply.yaml, light/medium/strict guidance…

    173 GitHub stars~945 tokensUpdated 6 days ago
    DevelopmentAuto-check passed
  • Tbd

    jlevy/strif

    Git-native issue tracking (beads), coding guidelines, knowledge injection, and spec-driven planning for AI agents.

    131 GitHub stars~3.5k tokensUpdated 4 mo ago
    DevelopmentAuto-check passed
  • Light Project Structure

    Light0305/Light-skills

    Audits, scaffolds and safely migrates research project folder structures, keeping existing repositories read-only until you approve exact moves from a plan.

    640 GitHub stars~3k tokensUpdated 3 mo ago
    DevelopmentAuto-check: notes
  • Claude Settings Audit

    getsentry/skills

    Official

    Analyze a repository to generate recommended Claude Code settings.json permissions.

    1k GitHub starsUsed in 4 repos~3k tokens
    DevelopmentAuto-check passed
  • Rocky Dev

    rocky-data/rocky

    Top-level router for Rocky development tasks. An agent skill from rocky-data/rocky.

    304 GitHub stars~2.1k tokensUpdated today
    DevelopmentAuto-check passed
  • Scaffold Project

    Marve10s/Better-Fullstack

    Scaffold a new app, API, backend, fullstack project, mobile app, polyglot service, monorepo, or starter with Better Fullstack.

    754 GitHub stars~716 tokensUpdated yesterday
    DevelopmentAuto-check passed

More from anam-org/metaxy

All 8 skills in this repo
  • Tach

    anam-org/metaxy

    This skill should be used when the user asks to "add a tach module", "configure tach layers", "define module boundaries", "set up interfaces", "run tach check", "check module boundaries", "tach…

    124 GitHub stars~1.2k tokensUpdated 8 days ago
    Auto-check passed
  • Claude Improve Config

    anam-org/metaxy

    Self-reflect on the current session to identify mistakes and propose improvements to .claude configuration (CLAUDE.md, hooks, skills).

    124 GitHub stars~1k tokensUpdated 8 days ago
    Auto-check passed
  • Docs Page Frontmatter

    anam-org/metaxy

    Write YAML front matter for documentation pages with appropriate titles and descriptions for social cards.

    124 GitHub stars~948 tokensUpdated 8 days ago
    Auto-check passed
  • Hypothesis

    anam-org/metaxy

    Use Hypothesis for property-based testing to automatically generate comprehensive test cases, find edge cases, and write more robust tests with minimal example shrinking.

    124 GitHub stars~1.7k tokensUpdated 8 days ago
    Auto-check passed
  • Narwhals

    anam-org/metaxy

    Effectively use Narwhals to write dataframe-agnostic code that works seamlessly across multiple Python dataframe libraries.

    124 GitHub stars~3.3k tokensUpdated 8 days ago
    Auto-check passed
  • Sybil

    anam-org/metaxy

    Use Sybil for testing code examples in documentation and docstrings.

    124 GitHub stars~1.6k tokensUpdated 8 days ago
    Auto-check passed

Works with

Categories

Questions about Metaxy

What does Metaxy do?

This skill should be used when the user asks to "define a feature", "create a BaseFeature class", "track feature versions", "set up metadata store", "field-level lineage", "FieldSpec", "FeatureDep"…. Metaxy is an agent skill from anam-org/metaxy.

When should I use Metaxy?

Metaxy fits situations like: asks to define a feature; create a BaseFeature class; track feature versions; set up metadata store.

How do I install Metaxy in Claude Code?

Run `npx skills add anam-org/metaxy --skill metaxy -a claude-code`. Or copy the skill folder (.claude-plugin/skills/metaxy in anam-org/metaxy) into .claude/skills/metaxy in your project. Claude Code loads it when a task matches its description.

How do I install Metaxy in Codex?

Run `npx skills add anam-org/metaxy --skill metaxy -a codex`. Or copy the skill folder (.claude-plugin/skills/metaxy in anam-org/metaxy) into .agents/skills/metaxy in your project. Codex loads it when a task matches its description.

Can I use Metaxy in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add anam-org/metaxy --skill metaxy -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/metaxy, .gemini/skills/metaxy, .github/skills/metaxy and .opencode/skills/metaxy in your project.

What does Metaxy need to run?

SKILL.md names no scripts, command-line tools or credentials: Metaxy is instructions for the agent only. Our summary lists: Python 3.

Does Metaxy access the network?

SKILL.md names 1 domain. As links in the text: docs.metaxy.io. This is read from the text; nothing was executed.

Is Metaxy safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Metaxy use?

Metaxy is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Metaxy use?

About 1.5k tokens (SKILL.md is roughly 6.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Metaxy?

Skills that share tags, products or a category with Metaxy: Deply Config (Vashkatsi/deply, 173 stars), Tbd (jlevy/strif, 131 stars), Light Project Structure (Light0305/Light-skills, 640 stars) and Claude Settings Audit (getsentry/skills, 1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Metaxy?

anam-org (a GitHub organization) maintains it in anam-org/metaxy, which has 124 GitHub stars. The repository holds 8 skills in this directory. The repository was last updated on September 30, 2026.

Source: anam-org/metaxy on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.