AGENT SKILL

mage-ai

Mage AI repo skill for platform operations, pipeline authoring, integrations, dbt, streaming, and AI workflows.

License 许可
Apache-2.0
最近更新
2026-09-03
Stars
★ 351
Forks
27

01 前言 Front Matter

name
mage-ai
license
Apache 2.0
metadata
{ "disco-role": "operating" }
description
Mage AI repo skill for platform operations, pipeline authoring, integrations, dbt, streaming, and AI workflows.
disable-model-invocation
true

02 资源文件 5

03 SKILL.md 正文

Mage AI

Use this skill for the mage-ai repository and package. It is self-contained: future work should rely on the bundled references and scripts in this skill tree, not on the original checkout.

Quick route map

User need Route
Install Mage, bootstrap a project, start the server, run a pipeline, clean logs/variables, or diagnose auth/logging/runtime startup issues sub-skills/platform-ops/
Author batch pipelines with Python, SQL, or R blocks; use runtime variables; work with dynamic blocks or templates sub-skills/pipeline-authoring/
Configure io_config.yaml, batch data integration pipelines, source/destination profiles, or connector settings sub-skills/batch-integrations/
Build or debug real-time pipelines, Kafka/CDC flows, or streaming sources/sinks sub-skills/streaming/
Add existing dbt projects, run dbt models or tests, or interpolate Mage values into dbt workflows sub-skills/dbt-workflows/
Set up AI generation, documentation helpers, or AI credentials and failures sub-skills/ai-workflows/

Shared facts

  • Public package name: mage-ai
  • Console entry point: mage
  • Python helper: mage_ai.run(pipeline_uuid, project_path=None, block_uuid=None, run_sensors=True, **global_vars)
  • Supported setup paths in the docs are Docker, pip, and conda.
  • Helpful extras: mage-ai[ai], mage-ai[streaming], and mage-ai[dbt].

How to use this skill

  1. Pick the route that matches the user's task.
  2. Read the route's SKILL.md and bundled references first.
  3. Use the bundled scripts for safe smoke checks when a task needs a quick validation.
  4. Keep execution, troubleshooting, and connector details inside the route that owns them.

Safe starting checks

  • scripts/smoke_cli.py — import and CLI help smoke test
  • scripts/check_project_layout.py — inspect a Mage project path without mutating it

What this skill does not do

  • It does not require the original repository checkout once the skill is loaded.
  • It does not execute cloud, broker, database, or model side effects unless the user explicitly asks for them.
  • It does not replace route-specific guidance with generic Mage documentation.

04 同仓库更多 Skills

VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-triton-opt

Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-solve

END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-skill

PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-pattern-fusion

Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-orchestrate

Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-feedback

Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).

★ 351