Skip to content

docs: add hosted managed agent sample to case indexes #88

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
danyalahmed1995 wants to merge 1 commit into
base: main
Choose a base branch
from
Open

docs: add hosted managed agent sample to case indexes #88

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion 09.Cases/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,4 +15,8 @@ This project demonstrates how to use Microsoft Foundry Local (e.g., Qwen2.5-1.5B
This narrative documents moving from concept to code through Microsoft Foundry and the Agent Framework. It uses a recruitment scenario with Recruiter and Applicant agents to illustrate a three-phase lifecycle: low-code workflow design in Foundry, code synchronization via the VS Code extension, and deployment on the Microsoft Agent Framework runtime using YAML-based “configuration as code.” The subdirectories contain the workflow YAML definitions and guidance for running the generated agents and workflows locally.

## [maf_harness_managed_agent](./maf_harness_managed_agent/README.md)
This project implements Anthropic's managed-agent pattern (“decoupling the brain from the hands”) with Microsoft Agent Framework and Microsoft Foundry. It separates a stateless `AgentHarness` (brain), durable append-only `SessionLog` (memory), and on-demand sandbox execution layer (hands), then demonstrates recovery after harness crashes via `wake(session_id)` and parallel orchestration with many-brains-many-hands workflows. The sample also includes Azure Functions/FastAPI hosting endpoints (`/sessions`, `/run`, `/events`, `/wake`, `/health`) and production-oriented extension points for replacing in-memory components with Azure services such as Cosmos DB, Blob Storage, Key Vault, and Azure Monitor.
This project implements Anthropic's managed-agent pattern (“decoupling the brain from the hands”) with Microsoft Agent Framework and Microsoft Foundry. It separates a stateless `AgentHarness` (brain), durable append-only `SessionLog` (memory), and on-demand sandbox execution layer (hands), then demonstrates recovery after harness crashes via `wake(session_id)` and parallel orchestration with many-brains-many-hands workflows. The sample also includes Azure Functions/FastAPI hosting endpoints (`/sessions`, `/run`, `/events`, `/wake`, `/health`) and production-oriented extension points for replacing in-memory components with Azure services such as Cosmos DB, Blob Storage, Key Vault, and Azure Monitor.

## [maf_harness_managed_hosted_agent](./maf_harness_managed_hosted_agent/README.md)
This sample shows how to build a Microsoft Foundry Hosted Agent with Microsoft Agent Framework and the Azure AI AgentServer SDK. It adapts the managed-agent harness pattern into a hosted deployment flow with separate layers for the agent brain, sandboxed hands, durable session log, and credential vault.
Use this sample to explore hosted-agent deployment with `azd`/`ai agent`, OpenAI Responses-compatible endpoints, sandboxed tool execution, durable JSONL session recovery, and production-oriented boundaries for credentials and external execution.
2 changes: 1 addition & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -182,4 +182,4 @@ If you encounter any issues or have questions:

---

**Start your journey with Microsoft Agent Framework today!** 🚀
**Start your journey with Microsoft Agent Framework today!** 🚀