mcp-request-idempotency-reference
Demonstrates the proposed idempotencyKey mechanism on tools/call with server-side deduplication and conflict fingerprint rejection. Includes a runnable client demo showing retry dedup and conflict behavior.
README
Request Idempotency — Reference Implementation
Minimal, runnable demonstration of the mechanism proposed in
SEP-3182: Request Idempotency:
an optional idempotencyKey argument on tools/call, with server-side
deduplication and explicit conflict-fingerprint semantics for a reused
key presented with different arguments.
This is a demonstration server, not a production implementation — the dedup store is an in-memory dict with no eviction. It exists to make the before/after failure mode concrete for reviewers, per the SEP process's prototype requirement ("a standalone proof-of-concept demonstrating the key mechanics").
Setup
pip install -r requirements.txt
Note on the version pin: this demo uses FastMCP from
mcp.server.fastmcp. As of the mcp SDK's 2.0.0 release, that class
was renamed and moved to MCPServer in mcp.server.mcpserver. Installing
a plain pip install mcp today will pull 2.0.0 and fail on import. The
pin above (mcp>=1.9.0,<2.0.0) avoids that; if you'd rather run against
the current SDK, swap the import for mcp.server.mcpserver.MCPServer
(same constructor/decorator surface for the pieces this demo uses).
Run
python3 client_demo.py
This spawns server.py as a stdio subprocess and runs three scenarios:
- No idempotency support — a lost-response retry double-charges.
- With
idempotencyKey— the same retry is deduplicated; the cached result is returned, no re-execution. - Conflict semantics — the same key reused with different arguments is rejected outright, not replayed or silently executed.
Verified working end-to-end against mcp==1.9.4 on 2026-08-01.
Note on the 2026-07-28 stateless spec: this demo's dedup logic
(_dedup_store in server.py) is keyed entirely by the client-supplied
idempotencyKey and never depends on protocol-level session state, so
nothing here needed to change when MCP went stateless. The one thing
worth calling out for anyone adapting this into a real server: in a
horizontally-scaled, stateless deployment, that store needs to be
shared across instances (a cache or database), not a per-process dict
like this demo uses — see the SEP's "Relationship to the 2026-07-28
stateless core and MRTR" section for why.
Files
server.py— the guarded (charge_guarded) and unguarded (charge_unguarded) tool implementations, plus areset_ledgertest helper.client_demo.py— drives all three scenarios againstserver.py.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。