One-day enterprise Agent workshop: skill testing, sandboxed multi-agent pipelines, Brain/Hands/Session/Vault hosted agent on Microsoft Foundry
概览
A hands-on workshop for engineering teams building . The samples in this repository walk you from a single-process skill-testing prototype, through multi-agent sandboxed pipelines, all the way to a hosted agent deployed on with full Azure infrastructure. Designed as a (≈ 6–8 hours of code + slides). The workshop is organized around four recurring engineering concerns that every enterprise Agent program eventually has to solve: A minimal project that tests a "business Agent" (an educational-video-script generator with a strict output template) against an . Two models — and — are run in parallel and compared in the console. - A with a deterministic output contract. - An that crafts edge-case prompts (single-turn and multi-turn). - A that checks the contract without an LLM. - An rubric scoring the response on 5 dimensions. - 10 curated edge-knowledge test cases. Use this module to teach the team how to before any infrastructure is involved.
README
以下为原文 README 节选，完整内容请查看原文。
Enterprise Agent Workshop
A hands-on workshop for engineering teams building production-grade Agents inside the enterprise. The samples in this repository walk you from a single-process skill-testing prototype, through multi-agent sandboxed pipelines, all the way to a hosted agent deployed on Microsoft Foundry with full Azure infrastructure.
Designed as a one-day workshop (≈ 6–8 hours of code + slides).
The workshop is organized around four recurring engineering concerns that every enterprise Agent program eventually has to solve:
| Concern | Where it shows up |
|---|---|
| Skill definition & adversarial testing | Skill_Testing_Agent/, AgentHarness_HostedAgent/ |
| Sandboxed multi-agent orchestration | OpenClaw_AgentHarness/, CodingAgent/ |
| Hosted-agent runtime & harness architecture (Brain / Hands / Session / Vault) | AgentHarness_HostedAgent/ |
| Cloud deployment to Microsoft Foundry | skill-testing-harness-deploy/ |
Workshop Modules
1. Skill_Testing_Agent/ — Adversarial Skill Testing 101
A minimal Microsoft Agent Framework + GitHub Copilot SDK (Python) project that tests a “business Agent” (an educational-video-script generator with a strict output template) against an adversarial test Agent. Two models — Claude Opus 4.7 and GPT-5.5 — are run in parallel and compared in the console.
Key concepts introduced:
- A business Agent with a deterministic output contract.
- An adversarial test Agent that crafts edge-case prompts (single-turn and multi-turn).
- A deterministic validator that checks the contract without an LLM.
- An LLM-as-judge rubric scoring the response on 5 dimensions.
- 10 curated edge-knowledge test cases.
Use this module to teach the team how to define a skill, write its evaluation, and reason about model differences before any infrastructure is involved.
2. OpenClaw_AgentHarness/ — Multi-Agent Sandbox Pipeline
A Docker-Compose stack that introduces OpenClaw as the sandbox / gateway. Three least-privilege Agents collaborate in a shared workspace through a self-closing loop:
| Agent | Role | Tool allowlist |
|---|---|---|
| A — Coder | Reads SPEC.md, writes solution.py |
read, write, edit |
| B — Tester | Reads spec + solution, writes test_solution.py |
read, write, edit |
| C — Runner | Runs pytest inside the sandbox, writes RUN_REPORT.md |
read, write, exec |
Only Agent C holds the exec capability, and it is constrained via tools.exec.allowedPaths to a single subdirectory. Models are served through GitHub Copilot’s built-in provider (Claude Opus 4.7).
推荐工具
换一个关键词，或者移除筛选条件。
安装
npx skillfish add kinfey/enterpriseagenticworkshop