---
id: agent-framework-declarative
version: "1.0.2"
license: unclear
license_treatment: permissive
maintenance: active
---

# agent-framework-declarative — Declarative specification support for Microsoft Agent Framework.

License: permissive · Maintenance: active · Downloads: 478.7K/mo

## What it is and what it does

This package adds declarative YAML-based specification support to Microsoft's Agent Framework, letting you define agents and workflows in configuration files rather than pure Python code. It wraps agent-framework-core with a higher-level abstraction layer and depends on httpx for HTTP operations, powerfx for formula evaluation, and pyyaml for parsing specifications.

The package splits its API into two stability tiers: WorkflowFactory and related workflow execution surfaces are production-ready, while AgentFactory and the YAML agent loading path are experimental and may change without notice—the package emits ExperimentalWarning when you use those symbols. This split means you can rely on declarative workflows in production but should treat agent configuration loading as subject to breaking changes.

Use it for:

- Define multi-step agent workflows in YAML and execute them without writing orchestration code
- Build agents with declarative configuration for easier version control and non-programmer configuration
- Prototype agent behavior by editing YAML specifications and reloading without code changes
- Integrate HTTP-based services into agent workflows using httpx under the hood
- Evaluate dynamic formulas within agent workflows using powerfx expressions

## Worth the install?

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

Enables building agents from YAML specifications using Microsoft's Agent Framework, with stable workflow support and experimental agent configuration loading.

Yes, if you are building agents within Microsoft's Agent Framework ecosystem and want YAML-based configuration. The stable workflow tier is production-ready, and the package has low install friction and active maintenance. However, avoid the experimental AgentFactory and YAML agent loading APIs in production code until they stabilize, as they carry no backward-compatibility guarantee.

## Install

pip install agent-framework-declarative
uv add agent-framework-declarative
poetry add agent-framework-declarative

## Installing agent-framework-declarative

Before you install: Low install friction with a pure Python wheel. Active maintenance as of 2026-08-14 with recent release. Depends on agent-framework-core, httpx, powerfx, and pyyaml.

License in practice: MIT license permits commercial and private use with minimal restrictions; permissive treatment means you can integrate freely into most projects.

Quickstart:

pip install agent-framework-declarative

from agent_framework_declarative import WorkflowFactory

workflow = WorkflowFactory.create(spec)
result = workflow.execute()

Requires Python 3.10 or later. Experimental agent loading APIs emit ExperimentalWarning and may break in future releases.

Verify before relying:

- Stability guarantees for the experimental AgentFactory and YAML agent loading path beyond the ExperimentalWarning
- Performance characteristics when loading large or complex YAML agent specifications
- Compatibility matrix with specific versions of agent-framework-core and powerfx

## Package facts

- License: not declared (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 478.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

declarative agent framework, yaml agent specification, agent framework declarative, workflow factory python, microsoft agent framework, declarative workflow builder, agent configuration yaml, agent-framework, declarative-config, workflow-orchestration

[View on SkillFed](https://skillfed.io/packages/agent-framework-declarative) · [View on PyPI](https://pypi.org/project/agent-framework-declarative/)