home.social

#modelcontextprotocol — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #modelcontextprotocol, aggregated by home.social.

  1. Security Flaws Exposed in Popular Database Projects' MCP Servers

    Critical security flaws have been uncovered in MCP servers used by popular analytics databases, leaving them vulnerable to risks like SQL injection and full database takeover due to faulty validation and authentication processes. These defects, discovered by Akamai security analyst Tomer Peled, highlight a…

    osintsights.com/security-flaws

    #VulnerabilityResearch #McpServers #DatabaseSecurity #AiApplications #ModelContextProtocol

  2. Anthropic's AI Tool Exposes to One-Click Remote Code Execution Risk

    A single click on Claude Code's generic dialog can unleash a major security risk, allowing an unsandboxed Node.js process to spawn with full user privileges. This vulnerability can be exploited using just two common JSON files, putting developers at risk of one-click remote code execution.

    osintsights.com/anthropics-ai-

    #RemoteCodeExecution #AiSecurity #Nodejs #ModelContextProtocol #OneclickExploit

  3. CodeGuardian is an MCP server that extends AI coding assistants with comprehensive code quality and security analysis.

    Developers can access enterprise-grade analysis directly in their AI assistant - reducing context switching and making secure coding easier to adopt.

    🔗 Read now: bit.ly/4u0VNhz

    #InfoQ #AI #ModelContextProtocol #AIagents #AIAssistedCoding

  4. CodeGuardian is an MCP server that extends AI coding assistants with comprehensive code quality and security analysis.

    Developers can access enterprise-grade analysis directly in their AI assistant - reducing context switching and making secure coding easier to adopt.

    🔗 Read now: bit.ly/4u0VNhz

    #InfoQ #AI #ModelContextProtocol #AIagents #AIAssistedCoding

  5. CodeGuardian is an MCP server that extends AI coding assistants with comprehensive code quality and security analysis.

    Developers can access enterprise-grade analysis directly in their AI assistant - reducing context switching and making secure coding easier to adopt.

    🔗 Read now: bit.ly/4u0VNhz

    #InfoQ #AI #ModelContextProtocol #AIagents #AIAssistedCoding

  6. CodeGuardian is an MCP server that extends AI coding assistants with comprehensive code quality and security analysis.

    Developers can access enterprise-grade analysis directly in their AI assistant - reducing context switching and making secure coding easier to adopt.

    🔗 Read now: bit.ly/4u0VNhz

    #InfoQ #AI #ModelContextProtocol #AIagents #AIAssistedCoding

  7. CodeGuardian is an MCP server that extends AI coding assistants with comprehensive code quality and security analysis.

    Developers can access enterprise-grade analysis directly in their AI assistant - reducing context switching and making secure coding easier to adopt.

    🔗 Read now: bit.ly/4u0VNhz

  8. Telefónica, Nokia work to speed adoption of APIs with agentic AI

    Their efforts include integrating the A2A protocol and MCP as a step towards building an ‘agent economy’ and…
    #Spain #ES #Europe #Europa #EU #Telefónica #Agent-to-AgentProtocol(A2A) #agenticAIecosystem #agenticeconomy #ModelContextProtocol(MCP) #Nokia
    europesays.com/spain/13198/

  9. Google veröffentlicht mit Deep Research und Deep Research Max zwei autonome KI-Agenten auf Basis von Gemini 3.1 Pro.

    Über das Model Context Protocol greifen die KI-Modelle auf interne Unternehmensdaten zu. Die Max-Variante generiert in asynchronen Analysen eigenständig Infografiken und schlägt den Vorgänger bei internen Evaluationen in 65,1 Prozent der Fälle.

    #Google #Gemini #KI #ModelContextProtocol #News
    all-ai.de/news/news26top/googl

  10. Google veröffentlicht mit Deep Research und Deep Research Max zwei autonome KI-Agenten auf Basis von Gemini 3.1 Pro.

    Über das Model Context Protocol greifen die KI-Modelle auf interne Unternehmensdaten zu. Die Max-Variante generiert in asynchronen Analysen eigenständig Infografiken und schlägt den Vorgänger bei internen Evaluationen in 65,1 Prozent der Fälle.

    #Google #Gemini #KI #ModelContextProtocol #News
    all-ai.de/news/news26top/googl

  11. Google veröffentlicht mit Deep Research und Deep Research Max zwei autonome KI-Agenten auf Basis von Gemini 3.1 Pro.

    Über das Model Context Protocol greifen die KI-Modelle auf interne Unternehmensdaten zu. Die Max-Variante generiert in asynchronen Analysen eigenständig Infografiken und schlägt den Vorgänger bei internen Evaluationen in 65,1 Prozent der Fälle.

    #Google #Gemini #KI #ModelContextProtocol #News
    all-ai.de/news/news26top/googl

  12. Cloudflare’s new #ModelContextProtocol (MCP) server powered by Code Mode enables #AIagents to interact with large APIs with minimal token usage.

    The server reduces context footprint across 2,500+ endpoints, improves multi-API orchestration, and provides a secure, code-centric execution environment for LLM agents.

    Deep dive on #InfoQbit.ly/4dTJgqQ

    #SoftwareArchitecture #LLMs #API

  13. Cloudflare’s new #ModelContextProtocol (MCP) server powered by Code Mode enables #AIagents to interact with large APIs with minimal token usage.

    The server reduces context footprint across 2,500+ endpoints, improves multi-API orchestration, and provides a secure, code-centric execution environment for LLM agents.

    Deep dive on #InfoQbit.ly/4dTJgqQ

    #SoftwareArchitecture #LLMs #API

  14. Cloudflare’s new #ModelContextProtocol (MCP) server powered by Code Mode enables #AIagents to interact with large APIs with minimal token usage.

    The server reduces context footprint across 2,500+ endpoints, improves multi-API orchestration, and provides a secure, code-centric execution environment for LLM agents.

    Deep dive on #InfoQbit.ly/4dTJgqQ

    #SoftwareArchitecture #LLMs #API

  15. Cloudflare’s new #ModelContextProtocol (MCP) server powered by Code Mode enables #AIagents to interact with large APIs with minimal token usage.

    The server reduces context footprint across 2,500+ endpoints, improves multi-API orchestration, and provides a secure, code-centric execution environment for LLM agents.

    Deep dive on #InfoQbit.ly/4dTJgqQ

    #SoftwareArchitecture #LLMs #API

  16. Cloudflare’s new (MCP) server powered by Code Mode enables to interact with large APIs with minimal token usage.

    The server reduces context footprint across 2,500+ endpoints, improves multi-API orchestration, and provides a secure, code-centric execution environment for LLM agents.

    Deep dive on bit.ly/4dTJgqQ

  17. Anthropic's MCP Flaw Exposes 200K Servers to Takeover Risk

    A security flaw in Anthropic's Model Context Protocol (MCP) could put a staggering 200,000 servers at risk of complete takeover, leaving thousands of machines vulnerable to attack. This design flaw, described as a vulnerability by security researchers, highlights a potentially disastrous weakness in a protocol meant to manage AI model…

    osintsights.com/anthropics-mcp

    #Anthropic #ModelContextProtocol #McpFlaw #AiSecurity #ServerTakeover

  18. Storybook 10.3 porta gli agenti AI dentro i componenti esistenti con il protocollo MCP

    Storybook 10.3 integra il Model Context Protocol per React, permettendo agli agenti AI di lavorare con componenti esistenti, eseguire test e correggersi autonomamente. Arrivano anche miglioramenti all'accessibilità e CSF Factories per Vue, Angular e Web Components.

    yoota.it/storybook-10-3-porta-

  19. Storybook 10.3 porta gli agenti AI dentro i componenti esistenti con il protocollo MCP

    Storybook 10.3 integra il Model Context Protocol per React, permettendo agli agenti AI di lavorare con componenti esistenti, eseguire test e correggersi autonomamente. Arrivano anche miglioramenti all'accessibilità e CSF Factories per Vue, Angular e Web Components.

    yoota.it/storybook-10-3-porta-

  20. @infoq Great case study. Central registry + human-in-the-loop approvals are exactly the right building blocks.

    We see the same pattern with ToolMesh: a gateway between agent and API. Instead of building an MCP server per API, a YAML file describes the endpoints — ToolMesh handles ACL, credential isolation, and audit trails.

    Pinterest's approach shows MCP works in production — when the governance layer is right.

    toolmesh.io

    #ModelContextProtocol #AIAgents #OpenSource

  21. @infoq Great case study. Central registry + human-in-the-loop approvals are exactly the right building blocks.

    We see the same pattern with ToolMesh: a gateway between agent and API. Instead of building an MCP server per API, a YAML file describes the endpoints — ToolMesh handles ACL, credential isolation, and audit trails.

    Pinterest's approach shows MCP works in production — when the governance layer is right.

    toolmesh.io

    #ModelContextProtocol #AIAgents #OpenSource

  22. @infoq Great case study. Central registry + human-in-the-loop approvals are exactly the right building blocks.

    We see the same pattern with ToolMesh: a gateway between agent and API. Instead of building an MCP server per API, a YAML file describes the endpoints — ToolMesh handles ACL, credential isolation, and audit trails.

    Pinterest's approach shows MCP works in production — when the governance layer is right.

    toolmesh.io

    #ModelContextProtocol #AIAgents #OpenSource

  23. @infoq Great case study. Central registry + human-in-the-loop approvals are exactly the right building blocks.

    We see the same pattern with ToolMesh: a gateway between agent and API. Instead of building an MCP server per API, a YAML file describes the endpoints — ToolMesh handles ACL, credential isolation, and audit trails.

    Pinterest's approach shows MCP works in production — when the governance layer is right.

    toolmesh.io

    #ModelContextProtocol #AIAgents #OpenSource

  24. @infoq Great case study. Central registry + human-in-the-loop approvals are exactly the right building blocks.

    We see the same pattern with ToolMesh: a gateway between agent and API. Instead of building an MCP server per API, a YAML file describes the endpoints — ToolMesh handles ACL, credential isolation, and audit trails.

    Pinterest's approach shows MCP works in production — when the governance layer is right.

    toolmesh.io

    #ModelContextProtocol #AIAgents #OpenSource

  25. Vor einem Jahr fragte @jwildeboer ob #MCP das Ende öffentlicher APIs bedeutet — weil Agenten unkontrolliert auf sie losgehen wie ein DDoS.

    Berechtigter Punkt. Die Lösung: ein Gateway zwischen Agent und API. Proaktives Rate Limiting, Retry-Budgets, Credential Isolation. Der Agent sieht nie den API-Key und kann nicht unkontrolliert feuern.

    Genau dafür haben wir ToolMesh gebaut. Open Source, self-hosted.

    toolmesh.io

    #ModelContextProtocol #OpenSource #AIAgents #SelfHosted

  26. Vor einem Jahr fragte @jwildeboer ob #MCP das Ende öffentlicher APIs bedeutet — weil Agenten unkontrolliert auf sie losgehen wie ein DDoS.

    Berechtigter Punkt. Die Lösung: ein Gateway zwischen Agent und API. Proaktives Rate Limiting, Retry-Budgets, Credential Isolation. Der Agent sieht nie den API-Key und kann nicht unkontrolliert feuern.

    Genau dafür haben wir ToolMesh gebaut. Open Source, self-hosted.

    toolmesh.io

    #ModelContextProtocol #OpenSource #AIAgents #SelfHosted

  27. Interne APIs sind einer der größten Bremsklötze für KI-Agenten im Unternehmen. Für die meisten wird es nie eine standardisierte KI-Schnittstelle (MCP) geben.

    Heute stellen wir ToolMesh vor — ein Open-Source-Gateway, das REST-APIs in Minuten anbindet: 30 Zeilen YAML ersetzen einen ganzen MCP-Server. ACL, Credential Isolation, Audit Trail und Output-Filter inklusive.

    16 APIs, 1.100+ Tools. Self-hosted, Apache 2.0.

    toolmesh.io

    #KI #MCP #ModelContextProtocol #OpenSource #AIAgents

  28. #Pinterest has deployed a production-ready #ModelContextProtocol (#MCP) ecosystem, enabling #AIagents to automate complex engineering tasks and integrate internal tools.

    Domain-specific MCP servers + central registry + human-in-the-loop approvals ⇒ boost security, governance & productivity - saving thousands of hours monthly.

    🔗 Details: bit.ly/4dVIhX6

    #InfoQ #SoftwareArchitecture

  29. #Pinterest has deployed a production-ready #ModelContextProtocol (#MCP) ecosystem, enabling #AIagents to automate complex engineering tasks and integrate internal tools.

    Domain-specific MCP servers + central registry + human-in-the-loop approvals ⇒ boost security, governance & productivity - saving thousands of hours monthly.

    🔗 Details: bit.ly/4dVIhX6

    #InfoQ #SoftwareArchitecture