Skills agent-framework
๐Ÿ“ฆ

agent-framework

Content revision r1 High Risk ๐Ÿ”‘ Env variablesโšก Contains scripts๐ŸŒ Network accessโš™๏ธ External commands

Build Microsoft Agent Framework Agents

Teams need repeatable patterns for building Agent Framework apps without missing server and debug setup. This skill guides Claude Code, Codex, and Claude through scaffolding agents, workflows, dependencies, and VS Code launch files.

Supports: Claude Codex Code(CC)
โš ๏ธ 38 Poor

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request
Review the Skillstore skill "agent-framework" from https://skillstore.io/skills/microsoft-agent-framework.md and its manifest at https://skillstore.io/api/skills/microsoft-agent-framework/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Test it

Using "agent-framework". Create a simple Foundry-backed agent.

Expected outcome:

  • A Python entrypoint using Microsoft Agent Framework and AzureAIClient.
  • A requirements list with pinned Agent Framework packages.
  • Environment placeholders for the Foundry project endpoint and model deployment name.

Using "agent-framework". Add debugging support for my agent.

Expected outcome:

  • VS Code task guidance for running the HTTP server under debugpy.
  • Launch configuration guidance for attaching the debugger.
  • Agent Inspector port settings that match the server command.

Using "agent-framework". Build a workflow with two agents.

Expected outcome:

  • Executor classes for each agent role.
  • A workflow graph with start executor and edges.
  • Streaming event handling for output and error events.

Security Audit

High Risk
v5 โ€ข 7/9/2026 Open versioned report

Most static findings are false positives caused by Markdown backticks, Python async syntax, VS Code task variables, and local loopback debug addresses. One configuration-handling concern remains: the skill tells agents to always create or update a .env file with real project values, which should require user confirmation and avoid overwriting existing secrets.

7
Files scanned
785
Lines analyzed
0
Review items
0
False positives ignored

Confirmed security concerns (1)

High
Environment file access
**ALWAYS create/update `.env` file**:
The skill explicitly instructs agents to always create or update a .env file with real project configuration. This is a legitimate setup step, but environment files can contain sensitive configuration and should be handled with user confirmation.
Audited by: claude View Audit History โ†’
Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report
Security Assessment

Copy report link

https://skillstore.io/skills/microsoft-agent-framework/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/microsoft-agent-framework/security.svg)](https://skillstore.io/skills/microsoft-agent-framework?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/microsoft-agent-framework?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/microsoft-agent-framework/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/microsoft-agent-framework.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Academic citations (APA ยท BibTeX ยท CFF)

APA citation

microsoft. (2026). agent-framework security audit report (audit version 5) [Author version unspecified]. Skillstore. https://skillstore.io/skills/microsoft-agent-framework/audits/5

BibTeX citation

@techreport{microsoft-microsoft-agent-framework-2026, author = {microsoft}, title = {agent-framework security audit report (audit version 5)}, institution = {Skillstore}, year = {2026}, number = {5}, url = {https://skillstore.io/skills/microsoft-agent-framework/audits/5}, note = {Author version unspecified} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "agent-framework security audit report (audit version 5)" version: "unspecified" type: report authors: - name: "microsoft" date-released: "2026-07-09" url: "https://skillstore.io/skills/microsoft-agent-framework/audits/5" identifiers: - type: other value: "skillstore:microsoft-agent-framework:audit:5" description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium
45
Architecture
85
Maintainability
85
Content
69
Community
83
Spec Compliance

What You Can Build

Create a Foundry-backed assistant

Scaffold a Python agent that connects to an Azure AI Foundry project and uses the recommended SDK imports.

Prototype a multi-agent workflow

Build a workflow where multiple chat agents exchange messages through Agent Framework executors.

Prepare local debugging

Add VS Code task and launch configuration guidance for Agent Inspector and debugpy workflows.

Try These Prompts

Create a basic agent
Create a Python Microsoft Agent Framework agent for my project. Use Azure AI Foundry and include the required environment placeholders.
Add MCP tools
Create an Agent Framework agent that can use MCP tools. Include the Microsoft Learn MCP reference and show where I can add more tools.
Build a workflow
Build a two-agent Microsoft Agent Framework workflow with separate writer and reviewer executors. Include streaming output handling.
Prepare server and debugging support
Convert this Agent Framework workflow into the HTTP server pattern and add VS Code debug tasks for Agent Inspector and debugpy.

Best Practices

  • Pin the Agent Framework package versions recommended by the skill before running generated projects.
  • Use virtual environments and venv-specific Python or pip commands for dependency installation.
  • Run the generated entrypoint once and fix startup errors before documenting the project.

Avoid

  • Do not use this skill for deployment workflows; use a deployment-specific skill instead.
  • Do not skip HTTP server or debug configuration unless the user asks for a minimal setup.
  • Do not place credentials in generated examples; use placeholders when real project values are unavailable.

Frequently Asked Questions

What does this skill create?
It creates guidance for Python agents, workflows, HTTP server wrappers, dependencies, and VS Code debugging support.
Does it deploy the agent?
No. It prepares code and local development setup, but deployment is out of scope.
Which SDK does it target?
It targets Microsoft Agent Framework for Python with Azure AI Foundry integration examples.
Can it build multi-agent workflows?
Yes. The references include executor-based workflow patterns, streaming events, and bidirectional agent loops.
Does it require Azure AI Foundry?
Most examples use Azure AI Foundry project endpoints and model deployments, so valid project configuration is expected.
Can Claude Code use this skill?
Yes. The report lists Claude Code, Claude, and Codex as supported tools.

Developer Details

Author

microsoft

License

MIT

Skillstore revision

r1

Version notice

The author did not declare a version.

Ref

1ffa7643651792ccb4bd3b15d924d2c97edff755

Maintenance freshness

7/18/2026

Usage

5 downloads ยท 125 views

File structure

๐Ÿ“ references/

๐Ÿ“„ agent-as-server.md

๐Ÿ“„ agent-samples.md

๐Ÿ“„ debug-setup.md

๐Ÿ“„ workflow-agents.md

๐Ÿ“„ workflow-basics.md

๐Ÿ“„ workflow-foundry.md

๐Ÿ“„ SKILL.md

More from microsoft

View all
View all