今日已更新 191 条资讯 | 累计 42602 条内容
关于我们

标签：#open

找到 2974 篇相关文章

AI 资讯

The repo that became its own good-first-issue

I've always loved teaching and helping others achieve things. There is a huge sense of accomplishment leveraging someone else's abilities. Not only telling them "you can!", but rather help them feel "I can". I was working as a developer for some time when I decided it was time to contribute to other projects. But, honestly, finding the right project, the right issue, and having the right timing was much harder than what I expected initially. I decided to create something that could help me out: scrape some orgs, find good-first-issue labels, and aggregate them together in a README file. Contributing to Open Source isn't easy for many reasons: The obvious, the technical: finding your first technically possible contribution is a combination of the right language, the right depth, and knowing which repo to search in. good-first-issues tackled that by scraping the language and the issue title to somewhat give me a little of context to start with. The not so obvious, the human side of it. My first contribution(s) were hard because I felt exposed, my weaknesses were in the wild for everyone to see and point them out to me. At least that was how I felt. And once I found the right issue, and I decided to give that step forward, many times I didn't get an answer back. That kills any motivation left. Probably PR ghosting is the biggest reason why someone quits their willingness to contribute to OSS. At first, good-first-issues was just a place others could come to find issues. But I realised it could be that issue. There were functionalities I wanted to bring and either I didn't have the time, or didn't have on the top of my head how to do them. "I can kill two birds with one stone" I thought. I knew where I wanted the repo to go, and I wanted it to be community-driven. Creating good self-contained, clear and approachable issues is an art in itself: I didn't want to create the obvious "Fix this typo" (which I did initially) or "Add your name as a contributor" issues. But I di

2026-06-05 原文 →
AI 资讯

The MCP SDK's EventStore Lives in Memory. Here's What Happens When Your Server Restarts.

I Built a Python Package to Fix SSE Resumability in the MCP SDK Your MCP server crashed. Your client reconnected. Every event from that session? Gone. The Gap The Model Context Protocol Python SDK ships with a built-in EventStore that powers SSE stream resumability — when a client reconnects with a Last-Event-ID header, the server replays the events it missed. This works great in development. The catch: that store lives entirely in memory. Restart the process, roll a new deployment, or — in a multi-worker setup — have the reconnecting client land on a different pod, and the session is gone. The store was local to the process that died. Resumability silently returns nothing. This isn't a bug in the SDK. It's a scope decision — the in-memory store is a correct, useful default for single-process development. But the moment you deploy to production, you need something durable. That's the gap mcp-persist fills. What It Does mcp-persist adds three drop-in EventStore backends — SQLite , Redis , and PostgreSQL — that survive process restarts and work across multi-worker deployments. Pick the one that fits your infrastructure; the API is identical across all three. pip install "mcp-persist[sqlite]" # no external service needed pip install "mcp-persist[redis]" # for multi-worker deployments pip install "mcp-persist[postgres]" # for teams already running Postgres The Two-Line Setup Wiring resumability by hand is tedious — you need a store, a StreamableHTTPSessionManager , a Starlette lifespan to open and close both, and a Mount . The with_persistence() helper collapses all of that. Pass your FastMCP instance, get back a runnable ASGI app: import uvicorn from mcp.server.fastmcp import FastMCP from mcp_persist import with_persistence mcp = FastMCP ( name = " MyServer " ) app = with_persistence ( mcp , backend = " sqlite " , url = " events.db " , ttl = 3600 ) uvicorn . run ( app , host = " 127.0.0.1 " , port = 8000 ) # MCP endpoint at /mcp Switching to Redis is a one-word change:

2026-06-05 原文 →
← 上一页 下一页 →