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:

265
active users

#mcp

14 posts12 participants0 posts today

🧠 What if your AI agent could learn to use your product, like a real teammate?

We just taught an AI agent how to use @Fleet's API through natural language, using MCP. No memorized commands. No rigid scripts. Just intelligent interaction.

It wasn’t all smooth sailing—we hit plenty of bumps.

👉 Read the full post to learn how MCP works, where we tripped up, and why it might be too early (or just early enough) to care.

victoronsoftware.com/posts/int
#AgenticAI #MCP #AI #DevTools

Victor on Software · Introducing MCP: Lessons from building an AI-driven serverHow we taught an AI agent to use our product with MCP

🚨 Breaking News: Yet another groundbreaking AI tool that promises to revolutionize code, all while being totally indispensable. 🙄 Meanwhile, #developers are still trying to figure out what "MCP" even stands for. 😂 Just another shiny object for the #GitHub magpies. 🐦✨
github.com/Klavis-AI/klavis #AItools #MCP #Innovation #HackerNews #ngated

Klavis AI (YC X25):  Open Source enterprise-grade MCP integration - Klavis-AI/klavis
GitHubGitHub - Klavis-AI/klavis: Klavis AI (YC X25): Open Source enterprise-grade MCP integrationKlavis AI (YC X25): Open Source enterprise-grade MCP integration - Klavis-AI/klavis

𝑰 𝒍𝒐𝒗𝒆 𝒕𝒉𝒊𝒔 𝒄𝒓𝒆𝒂𝒕𝒊𝒗𝒆 𝒘𝒂𝒚 𝒐𝒇 𝒖𝒔𝒊𝒏𝒈 𝑴𝑪𝑷: Instead of giving people a PDF to read, Jake exposes his work history and skills through a tiny web‑server that any AI chatbot can talk to linkedin.com/posts/shishs_intr

www.linkedin.comIntroduction | ShiSh S.MCP (https://lnkd.in/gFzVtMhK) is an open, plug‑and‑play standard that tells large‑language‑model apps (like #ChatGPT, #Claude, #Copilot, etc.) exactly how to pull in outside information or run small helper functions over ordinary HTTP: https://lnkd.in/gQeDBaZu Without a shared protocol, every time a developer wants an #LLM to look at a database, a résumé, a calendar, or a code repo, they have to write a one‑off integration. MCP replaces all those custom adapters with one common “language” that any model or data source can speak https://lnkd.in/gETKpGiS 𝑰 𝒍𝒐𝒗𝒆 𝒕𝒉𝒊𝒔 𝒄𝒓𝒆𝒂𝒕𝒊𝒗𝒆 𝒘𝒂𝒚 𝒐𝒇 𝒖𝒔𝒊𝒏𝒈 𝑴𝑪𝑷: Instead of giving people a PDF to read, Jake exposes his work history and skills through a tiny web‑server that any AI chatbot can talk to: https://ai.jakegaylor.com/. With #MCP in place, we’re just a step away from #AIAgents that carry our values, interests, needs, and schedules, then mingle with other agents to automatically uncover new business opportunities/collaborations for us. #AgenticAI #AIAgents #GenAI #AI
#MCP#AI#GenAI

#開源分享 直接在代碼編輯器裡查看Deepwiki內容的MCP:deepwiki-mcp,相當於內建了個文件瀏覽器,非官方的但比較實用

它通過MCP獲取Deepwiki的URL,抓取相關頁面,轉換為Markdown格式,並把結果返回給代碼編輯器

那你就可以直接在Cursor等中查看,不用再打開瀏覽器,即可即時獲取最新文件了

它可以抓取單頁或整個文件,能保持文件結構,並去除廣告/導航欄等無關內容

不需要註冊帳號或者API金鑰即可使用

專案地址: github.com/regenrek/deepwiki-mcp

⚠️ Major vulnerabilities found in MCP and A2A — two key AI agent frameworks 🧠🛠️

Researchers uncovered critical security issues in:
🔹 Anthropic’s Model Context Protocol (MCP)
🔹 Google’s Agent2Agent (A2A)

Threats include:
🧪 Tool poisoning — compromised functions warp agent behavior
🔓 Prompt injections — malicious inputs bypass safety
🤖 Rogue agents — faking capabilities to exploit systems

AI agent coordination is powerful — but without trust boundaries, it’s dangerous.

#AIsecurity #MCP #A2A #CyberRisk #LLMsecurity #AgenticAI
thehackernews.com/2025/04/expe

The Hacker NewsResearchers Demonstrate How MCP Prompt Injection Can Be Used for Both Attack and DefensePrompt injection flaws in Anthropic’s MCP and Google’s A2A protocols enable covert data exfiltration and AI manipulation.