---
id: microsoft-agents-activity
version: "1.3.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# microsoft-agents-activity — A protocol library for Microsoft Agents

License: permissive · Maintenance: active · Downloads: 937.6K/mo

## What it is and what it does

Microsoft Agents Activity is a protocol library that defines the core types and schemas for building conversational AI agents on Microsoft 365 platforms. It implements the Activity protocol—a messaging standard used across Teams, Copilot Studio, and web chat—providing Pydantic-validated data structures for different communication types (messages, typing indicators, conversation updates, events, invokes, and end-of-conversation signals). The library is the foundational layer of the Microsoft 365 Agents SDK; it does not itself host agents or manage conversation state, but rather supplies the type definitions that other packages in the SDK ecosystem (hosting, storage, authentication) build upon.

Developers use this package when they need to work with agent activities programmatically—parsing incoming messages, constructing outgoing responses with cards and attachments, or integrating custom logic into an agent pipeline. It is typically installed as a dependency of higher-level hosting packages, but can also be used standalone for activity validation or schema-driven development. The library requires Python 3.10 or later and depends only on pydantic.

Use it for:

- Define and validate incoming and outgoing agent messages when building a custom agent handler or middleware.
- Construct rich message cards (animations, attachments, interactive elements) to send from an agent to Teams or Copilot Studio.
- Parse activity types (message, typing, conversation update, invoke, event) to route and handle different communication patterns.
- Integrate a custom agent with the Microsoft 365 Agents SDK by using Activity types as the contract between your logic and the hosting layer.
- Validate agent payloads in a testing or integration layer without deploying a full hosting stack.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Provides core types and schemas implementing the Activity protocol for building conversational AI agents across Microsoft 365 platforms like Teams, Copilot Studio, and web chat.

Yes. This is a lightweight, actively maintained protocol library with no security vulnerabilities, permissive licensing, and a single well-known dependency. Install it if you are building agents for Microsoft 365 platforms or integrating with the Microsoft Agents SDK ecosystem. It is also safe to install as a transitive dependency of higher-level hosting packages.

## Install

pip install microsoft-agents-activity
uv add microsoft-agents-activity
poetry add microsoft-agents-activity

## Installing microsoft-agents-activity

Before you install: Low install friction with a single pure-Python dependency (pydantic). Active maintenance with a release 15 days ago and regular updates since first release in July 2025.

License in practice: MIT license permits free use, modification, and distribution with minimal restrictions—suitable for commercial and open-source projects.

Quickstart:

pip install microsoft-agents-activity

from microsoft_agents_activity import Activity, ActivityTypes

# Create a message activity
message = Activity(
    type=ActivityTypes.message,
    text="Hello from agent"
)

Requires Python 3.10 or later.

Verify before relying:

- Whether the Activity protocol implementation is fully compatible with all Microsoft 365 platforms mentioned or if some require additional hosting packages.
- Whether pydantic version constraints exist that might conflict with other dependencies in a larger project.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 937.6K/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

microsoft agents activity protocol, conversational ai types schemas, teams bot framework types, microsoft 365 agent messaging, activity protocol implementation, pydantic agent schemas, cross-platform chat agent types, microsoft-365, conversational-ai, protocol-types

[View on SkillFed](https://skillfed.io/packages/microsoft-agents-activity) · [View on PyPI](https://pypi.org/project/microsoft-agents-activity/)