by haasonsaas · MCP 服务器 · ★ 12
Agent Harness Agent Harness is a small Python adapter for running the same tool registry through OpenAI Agents SDK or Anthropic Claude Agent SDK provider implementations. The core package has no runtime SDK dependency; provider SDKs are imported lazily only when that provider is used. What It Provides A thread-safe global tool registry via . JSON Schema generation from Python type hints, including , , , , , and multi-type unions. A common and shape across providers. Lazy OpenAI and Claude provider adapters. Provider comparison helpers for running the same prompt against multiple adapters.
| Stars | 12 |
| Forks | 2 |
| Language | Python |
| Category | MCP 服务器 |
| License | MIT |
| Quality Score | 43.2/100 |
| Last Updated | 2025-10-07 |
| Created | 2025-10-06 |
| Platforms | claude-code, mcp, python |
| Est. Tokens | ~2k |
These tools work well together with agent-harness for enhanced workflows:
Explore other popular mcp 服务器 tools:
agent-harness is Unified harness for hot-swapping between OpenAI Agents SDK and Anthropic Claude Agent SDK with shared tool registry. It is categorized as a MCP 服务器 with 12 GitHub stars.
agent-harness is primarily written in Python. It covers topics such as agent-sdk, ai-agents, anthropic.
You can find installation instructions and usage details in the agent-harness GitHub repository at github.com/haasonsaas/agent-harness. The project has 12 stars and 2 forks, indicating an active community.
agent-harness is released under the MIT license, making it free to use and modify according to the license terms.