1. Habit
  2. Skills
  3. Bailian Managed Agent skill
modelstudioai · 9K installs

Bailian Managed Agent skill
Runs by itself, agents.yaml changed.

Agent infra changes stay reviewed.

Install the Bailian Managed Agent skill and make it a habit: follow https://gethabit.dev/skills/modelstudioai/bailian-managed-agent.md

  1. 1Copy
  2. 2Paste into Claude Code, Codex or Cursor
  3. 3It runs by itself

9K installs363 starsApache-2.0 licenceSource on GitHub ↗

Bailian Managed Agent

by modelstudioai · 9K installs

The skill

The bailian managed agent skill lets an agent manage hosted agents on Alibaba Cloud Bailian as code with bl managed-agent: init, validate, plan, apply and destroy from agents.yaml, plus sessions, files and memory stores.

Only when you ask.

The habit

Agent infra changes stay reviewed.

↻ Runs by itselfagents.yaml changed

How the habit works.
You don't have to.

Three questionsWhen does it start? How does it check its work? What does it remember?

01

It starts by itself

agents.yaml changed. Runs in GitHub Actions or Claude Code routine.

  • ↻ agents.yaml changed
  • push that changes agents.yaml

02

It checks its work

validate, plan, your yes. If a check fails, it stops.

  • ✓bl managed-agent validate passes
  • ✓bl managed-agent plan shows a diff before apply
  • ✓apply runs with --yes only after confirmation

03

It remembers

plan and apply log. The next run reads it first: the last applied plan and any failed creates.

loops/bailian-agents-yaml-loop/runs.md

One run, step by step.
Then it stops.

Stop ruleStops after one agents.yaml change and one apply per run.

  1. 01

    Read the protocol

    The agent reads the shared bailian-protocol file for the high-risk confirmation rules.

  2. 02

    Validate

    It runs bl managed-agent validate, which works offline.

  3. 03

    Plan

    It runs bl managed-agent plan and shows you the pending diff.

  4. 04

    Wait for your yes

    It asks for explicit confirmation and stops if you decline or do not answer.

  5. 05

    Apply

    After your yes it runs bl managed-agent apply --yes.

  6. 06

    Log

    It writes the plan and the apply result to the log.

What it may do.
Nothing more.

Safe by defaultIt writes only in its own folder and asks you before anything goes out.

Is it for you?
Use it when it fits.

Quick checkRead both lists. If the left one sounds like you, the Bailian Managed Agent skill helps.

Use it when

  • ✓ You create a hosted Bailian agent
  • ✓ You preview agents.yaml changes
  • ✓ You manage sessions or memory stores

Skip it when

  • – You call an already published app
  • – You manage your local agent's memory

Questions

Bailian Managed Agent skill, answered.

What is the difference between state rm and destroy?

state rm only edits the local state file; destroy deletes the remote resource.

Can it delete agents?

Only with destroy, and only after a separate explicit confirmation.

How do I install the Bailian Managed Agent skill?

Copy the prompt at the top into your agent. It installs the skill from modelstudioai/cli and sets up the habit. Or run npx skills add modelstudioai/cli --skill bailian-managed-agent.

Is the Bailian Managed Agent skill free?

Yes. It is published under Apache-2.0. Habit is free too.

Bailian Managed Agent. Now as a habit.

Install the Bailian Managed Agent skill and make it a habit: follow https://gethabit.dev/skills/modelstudioai/bailian-managed-agent.md

  • Free
  • No account
  • Undo in one line

All modelstudioai skills →More ops skills →

Works in

  • Claude Code
  • Codex
  • Cursor
  • ChatGPT
  • Claude app

Made by modelstudioai, published under Apache-2.0. Habit links to the original and never changes it. Is this your skill?

Copying didn't work here. Select the line and copy it by hand.