Clone Agent logo

Clone Agent

Organization
microsoft
clone-agent

Clone a Copilot Studio agent from the cloud. Guides through environment selection, agent selection, and downloads agent YAML files.

Overview

Publishermicrosoft
Repositoryskills-for-copilot-studio
Skill nameclone-agent
Stars
478
Forks
101
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by microsoft on GitHub. Read the source before you install it.

Installation

Install the Clone Agent AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/microsoft/skills-for-copilot-studio/tree/main/skills/clone-agent
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/microsoft/skills-for-copilot-studio.git /tmp/skills-for-copilot-studio
mkdir -p .claude/skills
cp -r /tmp/skills-for-copilot-studio/skills/clone-agent .claude/skills/clone-agent
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Clone Agent in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Clone Agent on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Clone Agent is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Clone Agent

Guided flow to clone a Copilot Studio agent from the cloud to a local workspace. Walks the user through environment selection, agent selection, and downloads the agent's YAML files.

IMPORTANT: Do Not Modify Scripts

This is a new capability under active development. If the manage-agent script fails, do not attempt to fix, patch, or work around it. Instead:

  1. Show the user the full error output
  2. Direct them to https://github.com/microsoft/skills-for-copilot-studio/issues
  3. Suggest using the Copilot Studio VS Code extension directly as a fallback for push/pull/clone

Phase 0: Resolve Configuration

Look for existing connection details in this order. Stop at the first source that provides a tenantId:

0a. Scan for existing .mcs/conn.json files

Search the filesystem for previously cloned agents — these already have all the connection details:

Glob: **/.mcs/conn.json

Also search common project locations:

Glob: /Users/**/projects/**/.mcs/conn.json

Each conn.json contains TenantId, EnvironmentId, DataverseEndpoint, AgentManagementEndpoint, and AccountInfo. Parse any found files and collect unique tenant/environment pairs. If matches are found, present them to the user:

Found existing agent connections:

  1. Employee Assistant — Elad's Env (tenant: 8a235459...)
  2. IT Agent — Contoso Dev (tenant: efb073bb...)

Use one of these, or enter a different tenant ID?

If the user picks one → extract tenantId, environmentId, environmentUrl, agentMgmtUrl and skip to Phase 2 (or Phase 1 if they want a different environment).

0b. Check for a Copilot Studio URL

If the user pastes a Copilot Studio URL (e.g. from their browser while viewing the agent), extract environment and agent IDs directly — no need for list-envs or list-agents.

Recognised URL formats:

https://copilotstudio.microsoft.com/environments/<envId>/bots/<agentId>/overview
https://copilotstudio.preview.microsoft.com/environments/<envId>/bots/<agentId>

If the user provides a URL like this, you still need a tenantId (from a conn.json or by asking the user). Then skip to Phase 3, passing --url instead of --agent-id / --environment-id / --environment-url / --agent-mgmt-url:

bash
node ${CLAUDE_SKILL_DIR}/../../scripts/manage-agent.bundle.js clone \
  --workspace "." \
  --tenant-id "<tenantId>" \
  --url "<copilotStudioUrl>"

The script will parse the URL to extract the environment ID and agent ID, then resolve the remaining environment details (Dataverse URL, agent management URL) automatically via the BAP API.

0c. Ask the user

If no conn.json found (or user wants a different tenant) → ask for their tenant ID (required).

No --client-id is needed — the script uses VS Code's first-party client ID with interactive browser login automatically.

Phase 1: Select Environment

Tell the user: "A browser window may open for Microsoft sign-in (tokens are cached after first login)."

Run the list-envs command:

bash
node ${CLAUDE_SKILL_DIR}/../../scripts/manage-agent.bundle.js list-envs \
  --tenant-id "<tenantId>"

Timeout: 300000ms (5 minutes) — set this on the Bash tool call to allow time for browser auth.

Parse the JSON output. On success (status: "ok"), present the environments as a numbered list to the user:

1. Contoso Dev (abc123-...)
2. Contoso Prod (def456-...)

Ask the user to pick an environment. Extract:

  • environmentId
  • environmentUrl (the url or dataverseUrl field)
  • agentMgmtUrl (the agentManagementUrl field)

Phase 2: Select Agent

First, ask the user: "Do you want to list only agents you own, or all agents in the environment?"

  • If only mine → run without --no-owner (default, filters by current user)
  • If all agents → add --no-owner flag
bash
node ${CLAUDE_SKILL_DIR}/../../scripts/manage-agent.bundle.js list-agents \
  --tenant-id "<tenantId>" \
  --environment-url "<environmentUrl>" \
  [--no-owner]

Timeout: 300000ms (5 minutes)

Parse the JSON output. Each agent has agentId, displayName, and ownedByCurrentUser. Present agents as a numbered list:

1. My Support Bot (yours)
2. HR Assistant
3. IT Helpdesk (yours)

Ask the user to pick an agent. Extract the agentId.

Phase 3: Clone

Run the clone command. The workspace should be the current directory (.). The --agent-id is required (from Phase 2):

bash
node ${CLAUDE_SKILL_DIR}/../../scripts/manage-agent.bundle.js clone \
  --workspace "." \
  --tenant-id "<tenantId>" \
  --agent-id "<agentId>" \
  --environment-id "<environmentId>" \
  --environment-url "<environmentUrl>" \
  --agent-mgmt-url "<agentMgmtUrl>"

Timeout: 300000ms (5 minutes)

On success:

  1. Run Glob: **/agent.mcs.yml to find and show the cloned agent file
  2. Check that .mcs/conn.json was created in the agent directory
  3. Summarize what was cloned and where the files are

Error Handling

ErrorResolution
Browser auth fails or times outAsk user to try again; check that they have access to the tenant
No environments foundVerify tenant ID is correct and user has Power Platform access
No agents foundVerify environment selection; user may need Copilot Studio access
Clone failsCheck permissions; the user needs maker/admin access to the agent

Frequently asked questions

What does the Clone Agent AI skill do?

Clone a Copilot Studio agent from the cloud. Guides through environment selection, agent selection, and downloads agent YAML files.

Why use Clone Agent on TypingMind?

Because you install it once and use it with any model. Clone Agent is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Clone Agent in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/microsoft/skills-for-copilot-studio/tree/main/skills/clone-agent. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Clone Agent?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Clone Agent?

As many as you like. As long as a model supports skills, you can use Clone Agent with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Clone Agent AI skill free?

Yes. It is published on GitHub by microsoft under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇