veganism.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
Veganism Social is a welcoming space on the internet for vegans to connect and engage with the broader decentralized social media community.

Administered by:

Server stats:

295
active users

#mcp

20 posts19 participants1 post today
.:\dGh/:.<p>The writing was on the wall the moment <a href="https://mastodon.social/tags/Laravel" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Laravel</span></a> raised USD $60M from investments.</p><p>Eventually they would have to shove in <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> because of shareholders.</p><p><a href="https://laravel-news.com/laravel-boost-your-ai-coding-starter-kit" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">laravel-news.com/laravel-boost</span><span class="invisible">-your-ai-coding-starter-kit</span></a></p><p><a href="https://mastodon.social/tags/PHP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>PHP</span></a> <a href="https://mastodon.social/tags/Programming" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Programming</span></a> <a href="https://mastodon.social/tags/ArtificialIntelligence" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ArtificialIntelligence</span></a> <a href="https://mastodon.social/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> <a href="https://mastodon.social/tags/LaravelBoost" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LaravelBoost</span></a> <a href="https://mastodon.social/tags/Claude" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Claude</span></a> <a href="https://mastodon.social/tags/Copilot" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Copilot</span></a> <a href="https://mastodon.social/tags/Cursor" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Cursor</span></a> <a href="https://mastodon.social/tags/Junie" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Junie</span></a> <a href="https://mastodon.social/tags/SoftwareDevelopment" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>SoftwareDevelopment</span></a> <a href="https://mastodon.social/tags/WebDevelopment" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>WebDevelopment</span></a> <a href="https://mastodon.social/tags/Development" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Development</span></a> <a href="https://mastodon.social/tags/WebDev" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>WebDev</span></a> <a href="https://mastodon.social/tags/Coding" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Coding</span></a> <a href="https://mastodon.social/tags/Code" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Code</span></a></p>
梅伊娜<p>MCP 忽视了分布式系统中来之不易的教训</p><p>「人工智能行业面临着一个选择:要么从 RPC 四十年的演变中吸取教训,要么重蹈覆辙,重蹈覆辙。按照目前的发展轨迹,关键功能都是事后才添加的,我们选择重复,而企业将为完全可以避免的部署失败、安全漏洞和运营噩梦付出代价。」</p><p><a href="https://news.ycombinator.com/item?id=44846871" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">news.ycombinator.com/item?id=4</span><span class="invisible">4846871</span></a></p><p><a href="https://mastodon.social/tags/%E6%88%91%E5%9C%A8%E7%9C%8B%E4%BB%80%E4%B9%88" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>我在看什么</span></a> <a href="https://mastodon.social/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a></p>
Hassan Habib<p>Ever wish your AI could do more than just chat?<br>I connected my offline AI model to the Model Context Protocol (MCP) so it can call tools, run commands, and fetch info without touching the internet.</p><p>✅ 100% private<br>✅ Works with your own tools<br>✅ Runs locally</p><p>If you’ve got an AI at home… it’s time to make it work for you.<br>📺 Watch the full video here: <a href="https://www.youtube.com/watch?v=G-0HP2yHnLM" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="">youtube.com/watch?v=G-0HP2yHnLM</span><span class="invisible"></span></a></p><p><a href="https://techhub.social/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://techhub.social/tags/chatgpt" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>chatgpt</span></a> <a href="https://techhub.social/tags/offline" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>offline</span></a> <a href="https://techhub.social/tags/mcp" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>mcp</span></a></p>
Tedi Heriyanto<p>The double-edged sword of MCP: Understanding the threat landscape for AI workflows: <a href="https://redcanary.com/blog/threat-detection/mcp-ai-workflows/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">redcanary.com/blog/threat-dete</span><span class="invisible">ction/mcp-ai-workflows/</span></a></p><p><a href="https://infosec.exchange/tags/threatdetection" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>threatdetection</span></a> <a href="https://infosec.exchange/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ai</span></a> <a href="https://infosec.exchange/tags/mcp" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>mcp</span></a></p>
Veit Schiele 🔜 @FrOSCon<p>A malicious Jira ticket can cause Cursor to exfiltrate secrets from the repository or local file system. But this is not just a problem with Cursor: GitHub MCP connections can also be exploited to expose private repository data, and a vulnerability in GitLab Duo allowed private information to be exposed through automatically rendered HTML code.<br><a href="https://labs.zenity.io/p/when-a-jira-ticket-can-steal-your-secrets" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">labs.zenity.io/p/when-a-jira-t</span><span class="invisible">icket-can-steal-your-secrets</span></a><br><a href="https://mastodon.social/tags/Jira" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Jira</span></a> <a href="https://mastodon.social/tags/Cursor" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Cursor</span></a> <a href="https://mastodon.social/tags/GitHub" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>GitHub</span></a> <a href="https://mastodon.social/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> <a href="https://mastodon.social/tags/GitLab" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>GitLab</span></a> <a href="https://mastodon.social/tags/ITSecurity" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ITSecurity</span></a> <a href="https://mastodon.social/tags/Vulnerability" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Vulnerability</span></a></p>
RolandRides<p>Es ist einfach wunderbar, aber niemand versteht mich ;-) <a href="https://mastodon.social/tags/ki" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ki</span></a> <a href="https://mastodon.social/tags/genai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>genai</span></a> <a href="https://mastodon.social/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://mastodon.social/tags/mcp" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>mcp</span></a></p>
Lars Marowsky-Brée 😷<p>On the subject of <a href="https://mastodon.online/tags/LLMs" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLMs</span></a> in tech, there's one aspect that I find highly entertaining: </p><p>Complete, correct, unambiguous, uptodate, in short: great documentation?</p><p>Has become a mandatory product requirement. </p><p>Because that's how you enable your software to be "AI-first".</p><p>I'll take it. </p><p><a href="https://mastodon.online/tags/SoftwareEngineering" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>SoftwareEngineering</span></a> <a href="https://mastodon.online/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> <a href="https://mastodon.online/tags/AIAssistant" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AIAssistant</span></a></p>
Annexus Technologies<p>Agentic AI is growing fast—but most MCP setups aren’t built for real-world scale or security.<br>Cato’s new MCPSaaS framework fixes that with strong security, easy scaling, and better user experience—ready for the next wave of AI.</p><p><a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://mastodon.social/tags/Cybersecurity" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Cybersecurity</span></a> <a href="https://mastodon.social/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> <a href="https://mastodon.social/tags/CatoNetworks" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>CatoNetworks</span></a> </p><p><a href="https://www.catonetworks.com/blog/designing-the-future-of-agentic-ai/?utm_source=mastodon&amp;utm_medium=Zoho+Social" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">catonetworks.com/blog/designin</span><span class="invisible">g-the-future-of-agentic-ai/?utm_source=mastodon&amp;utm_medium=Zoho+Social</span></a></p>
Mads B. Klinkby 🌻<p>Amazed by <a href="https://fosstodon.org/tags/Playwright" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Playwright</span></a> testing framework's <a href="https://fosstodon.org/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> server for <a href="https://fosstodon.org/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a>.<br>The efficiency factor is just crazy 🤯compared to hand coding or maintaining :javascript: system test cases!<br>For my scenario GPT-5 produced better code than Claude4 Sonnet.</p>
Docker Blog<p>MCP Horror Stories: The Supply Chain Attack<br><a href="https://techhub.social/tags/Products" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Products</span></a> <a href="https://techhub.social/tags/Docker" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Docker</span></a> <a href="https://techhub.social/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> <a href="https://techhub.social/tags/MCPserver" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCPserver</span></a> <a href="https://techhub.social/tags/Security" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Security</span></a></p><p><a href="https://www.docker.com/blog/mcp-horror-stories-the-supply-chain-attack/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">docker.com/blog/mcp-horror-sto</span><span class="invisible">ries-the-supply-chain-attack/</span></a></p>
GripNews<p>🌘 你的大型語言模型並不在乎 MCP<br>➤ 解碼大型語言模型與工具調用的實質聯繫<br>✤ <a href="https://hackteam.io/blog/your-llm-does-not-care-about-mcp/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">hackteam.io/blog/your-llm-does</span><span class="invisible">-not-care-about-mcp/</span></a><br>本文闡述,儘管模型上下文協議(MCP)已成為建立AI代理時進行工具調用的標準,但大型語言模型(LLM)本身並不需要理解MCP的運作細節。文章解釋了LLM如何透過工具調用獲取外部資訊,以及MCP如何為開發者簡化整合眾多工具的複雜性。重點在於MCP是為開發者服務,以便更高效地進行上下文工程,而非直接影響LLM的理解能力。<br>+ 原來LLM對MCP毫無感知,這項標準完全是為了開發者便利!<br>+ MCP就像是AI工具界的USB-C,統一介面,省去不少客製化麻煩。<br><a href="https://mastodon.social/tags/%E5%A4%A7%E5%9E%8B%E8%AA%9E%E8%A8%80%E6%A8%A1%E5%9E%8B" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>大型語言模型</span></a> <a href="https://mastodon.social/tags/%E5%B7%A5%E5%85%B7%E8%AA%BF%E7%94%A8" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>工具調用</span></a> <a href="https://mastodon.social/tags/%E4%B8%8A%E4%B8%8B%E6%96%87%E5%B7%A5%E7%A8%8B" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>上下文工程</span></a> <a href="https://mastodon.social/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> <a href="https://mastodon.social/tags/AI%E4%BB%A3%E7%90%86" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI代理</span></a></p>
publictorsten<p>Jedes mal, wenn ich "Model Context Protocol" lese, denke ich an Tron.<a href="https://mastodon.social/tags/mcp" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>mcp</span></a></p>
Christoffer S.<p>I can't quite put my finger on it, but this post by Daniel Miessler about his personal MCP-exposed "Daemon" is actually pretty cool.</p><p><a href="https://danielmiessler.com/blog/launching-daemon-personal-api" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">danielmiessler.com/blog/launch</span><span class="invisible">ing-daemon-personal-api</span></a></p><p>The thing I like is that you choose what you expose, in a machine readable format. </p><p>I don't know... it just tickles me a little bit.</p><p><a href="https://swecyb.com/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://swecyb.com/tags/Future" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Future</span></a> <a href="https://swecyb.com/tags/API" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>API</span></a> <a href="https://swecyb.com/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a></p>
RolandRides<p>Immer wenns um <a href="https://mastodon.social/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> geht bei <a href="https://mastodon.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> denke ich xzibit. Yo Dawg I herd you like…</p>
squillace<p>yeah, yeah, AI this AI that... but **some of it** is very useful. But you want to build your <a href="https://hachyderm.io/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> servers with fixed and immutable tooling -- and you want those tools to be discoverable on the fly, with capability controls as well. </p><p>Yeah, it's <a href="https://hachyderm.io/tags/webassembly" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>webassembly</span></a> components, but it does the right thing.</p>
deadprogram<p>Skynet is one of the featured projects on Hackster right now. Very cool!</p><p><a href="https://www.hackster.io/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="">hackster.io/</span><span class="invisible"></span></a></p><p><a href="https://social.tinygo.org/tags/skynet" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>skynet</span></a> <a href="https://social.tinygo.org/tags/drones" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>drones</span></a> <a href="https://social.tinygo.org/tags/robots" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>robots</span></a> <a href="https://social.tinygo.org/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://social.tinygo.org/tags/mcp" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>mcp</span></a> <a href="https://social.tinygo.org/tags/docker" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>docker</span></a> <a href="https://social.tinygo.org/tags/ollama" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ollama</span></a> <a href="https://social.tinygo.org/tags/tinygo" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>tinygo</span></a> <a href="https://social.tinygo.org/tags/golang" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>golang</span></a></p>
january1073<p>The widespread MCP creates a large number of potential entry points for attacks on users' systems. Docker has published an overview of this:<br><a href="https://www.docker.com/blog/mcp-security-issues-threatening-ai-infrastructure/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">docker.com/blog/mcp-security-i</span><span class="invisible">ssues-threatening-ai-infrastructure/</span></a><br><a href="https://infosec.exchange/tags/mcp" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>mcp</span></a> <a href="https://infosec.exchange/tags/agenticai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>agenticai</span></a> <a href="https://infosec.exchange/tags/security" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>security</span></a></p>
James Montemagno<p>That 128 tool limit is GONE in vs code! Virtual tools just landed and now give you huge flexibility to add more <a href="https://mastodon.social/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> servers and more extensions to extend agent mode. I break it all down:</p><p><a href="https://youtu.be/qHl_KBUyid0" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">youtu.be/qHl_KBUyid0</span><span class="invisible"></span></a></p><p><a href="https://mastodon.social/tags/githubcopilot" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>githubcopilot</span></a> <a href="https://mastodon.social/tags/vscode" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>vscode</span></a></p>
Karlheinz Agsteiner<p>Maybe a bit of an explanation.</p><p>First, I'm deeply in the pro AI camp, spend much of my private life tinkering with Cursor.</p><p>And still...</p><p>MCP looks like this:</p><p>You --&gt; LLM --&gt; Client software --&gt; Server software.</p><p>MCP connects all kinds of servers that can do very powerful things like wipe your harddisk or book a flight, or that pretend to do powerful things while actually doing malicous stuff on one side with LLMs that can hallucinate, that can be convinced to execute malicious said malicious stuff etc on the other side.</p><p>And MCP just shrugs about this, instead of offering hard-coded security mechanisms.</p><p>From their own spec:</p><p>"Implementation Guidelines<br>While MCP itself cannot enforce these security principles at the protocol level, implementors SHOULD:</p><p>&lt;list of things the client layer should do so that the user is protected from whatever the LLM hallucinates or the malicous server tries to convince it to do.&gt;"</p><p>Compare that shit to e.g. all the sandboxes that OpenAI built into their agent framework. That's how you build reliable software with LLMs.</p><p>Okay, done ranting 🙂 </p><p><a href="https://hachyderm.io/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>llm</span></a> <a href="https://hachyderm.io/tags/mcp" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>mcp</span></a> <a href="https://hachyderm.io/tags/openai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>openai</span></a> <a href="https://hachyderm.io/tags/anthropic" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>anthropic</span></a> <a href="https://hachyderm.io/tags/security" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>security</span></a></p>
Karlheinz Agsteiner<p>Okay, I have to write this.</p><p>MCP is the single most dangerous security risk in all histtory of software. </p><p>It's a really dumb and way too simplistic protocol. But easy to use so it spreads.</p><p>So, there, I said it.</p><p><a href="https://hachyderm.io/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>MCP</span></a> <a href="https://hachyderm.io/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> <a href="https://hachyderm.io/tags/agent" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>agent</span></a></p>