home.social

#model-context-protocol — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #model-context-protocol, aggregated by home.social.

fetched live
  1. Released mcp-wire 0.3.0 — a small Go CLI that installs and configures MCP servers across Claude Code, Codex, and OpenCode from one interface.

    New in this version:
    - `mcp-wire doctor` diagnostics
    - richer TUI metadata
    - `mcp-wire metadata` for agents/scripts
    - GitHub, Notion, and Linear in the curated catalog
    - safer credential storage

    andreagrandi.it/posts/announci

    #mcp #ModelContextProtocol #cli #opensource #golang

  2. Released mcp-wire 0.3.0 — a small Go CLI that installs and configures MCP servers across Claude Code, Codex, and OpenCode from one interface.

    New in this version:
    - `mcp-wire doctor` diagnostics
    - richer TUI metadata
    - `mcp-wire metadata` for agents/scripts
    - GitHub, Notion, and Linear in the curated catalog
    - safer credential storage

    andreagrandi.it/posts/announci

    #mcp #ModelContextProtocol #cli #opensource #golang

  3. Tool poisoning turns an AI agent's most trusted input, its tool descriptions, into an attack surface. How MCP poisoning works and how to defend against it. hackernoon.com/your-ai-agent-t #modelcontextprotocol

  4. Tool poisoning turns an AI agent's most trusted input, its tool descriptions, into an attack surface. How MCP poisoning works and how to defend against it. hackernoon.com/your-ai-agent-t #modelcontextprotocol

  5. Discover how #GitHub reduced token usage by up to 62% in the agentic workflows running across their own repositories.

    The team achieved these efficiencies by:
    ➡️ Pruning unused #ModelContextProtocol (MCP) tools
    ➡️ Replacing MCP calls with native GitHub CLI invocations
    ➡️ Introducing daily audit and optimization agents

    🔗 Learn more: bit.ly/4vefjai

    #AI #AgenticAI #InfoQ

  6. Discover how reduced token usage by up to 62% in the agentic workflows running across their own repositories.

    The team achieved these efficiencies by:
    ➡️ Pruning unused (MCP) tools
    ➡️ Replacing MCP calls with native GitHub CLI invocations
    ➡️ Introducing daily audit and optimization agents

    🔗 Learn more: bit.ly/4vefjai

  7. The AI agent "Wild West" is over. It’s time for engineering discipline. 🤠❌

    Giving an LLM access to every API, database, and MCP server and hoping it figures things out leads to chaos, hallucinations, and burnt tokens.

    To fix this, I built a tool harness: a proxy-driven, benchmark-validated ecosystem to make environments truly "AI-ready."

    Key takeaways from my latest article:
    1/ MCP is a transport layer, not a UX. Expose high-level intents, not raw, complex REST endpoints.
    2/ Skills need a context budget. Keep instructions tight and load them on-demand so the model doesn't lose focus.
    3/ Implement a Learning Loop. Use a transparent proxy to capture interactions, write behavioral unit tests for your AI with strict assertions, and optimize skills based on real failures and token costs.

    Treat your tool harness with the same discipline you treat your test suite.

    Read the full post here:
    👉 blog.mornati.net/your-ai-agent

    #AI #LLM #MCP #ModelContextProtocol #Architecture #SoftwareEngineering

  8. The AI agent "Wild West" is over. It’s time for engineering discipline. 🤠❌

    Giving an LLM access to every API, database, and MCP server and hoping it figures things out leads to chaos, hallucinations, and burnt tokens.

    To fix this, I built a tool harness: a proxy-driven, benchmark-validated ecosystem to make environments truly "AI-ready."

    Key takeaways from my latest article:
    1/ MCP is a transport layer, not a UX. Expose high-level intents, not raw, complex REST endpoints.
    2/ Skills need a context budget. Keep instructions tight and load them on-demand so the model doesn't lose focus.
    3/ Implement a Learning Loop. Use a transparent proxy to capture interactions, write behavioral unit tests for your AI with strict assertions, and optimize skills based on real failures and token costs.

    Treat your tool harness with the same discipline you treat your test suite.

    Read the full post here:
    👉 blog.mornati.net/your-ai-agent

  9. CoinBase releases "Base MCP" an integration that enables users to connect their Base Account to AI clients including ChatGPT, Claude and Cursor using the Model Context Protocol (MCP). Users can ask the agent to send funds, swap tokens, check balances, review transaction history and interact with DeFi applications. coindesk.com/tech/2026/05/26/c #Crypto #CoinBase #BaseMCP #AI #AIAgents #ChatGPT #Claude #Cursor #Integration #MCP #ModelContextProtocol #CryptoTrading

  10. CoinBase releases "Base MCP" an integration that enables users to connect their Base Account to AI clients including ChatGPT, Claude and Cursor using the Model Context Protocol (MCP). Users can ask the agent to send funds, swap tokens, check balances, review transaction history and interact with DeFi applications. coindesk.com/tech/2026/05/26/c

  11. 🤖 Testing & observing AI agents shouldn't be an afterthought. If you wait until production to see how your agents behave, you're playing a risky game.

    3 expert speakers will demonstrate how to build robust agentic apps using Spring AI, and show you how to take control of the chaos.

    🔗 Save your seat today: linkedin.com/events/7460341650

    #AIagents #SoftwareTesting #SpringAI #OpenTelemetry #ModelContextProtocol

  12. 🤖 Testing & observing AI agents shouldn't be an afterthought. If you wait until production to see how your agents behave, you're playing a risky game.

    3 expert speakers will demonstrate how to build robust agentic apps using Spring AI, and show you how to take control of the chaos.

    🔗 Save your seat today: linkedin.com/events/7460341650

    #AIagents #SoftwareTesting #SpringAI #OpenTelemetry #ModelContextProtocol

  13. #AWS has made its managed #ModelContextProtocol (MCP) server generally available, giving AI coding agents controlled access to AWS APIs, documentation & operational workflows through a standard interface.

    It offers a safer, more auditable way to connect AI agents to AWS services without exposing broad credentials.

    Learn more: bit.ly/49PyxL6

    #CloudComputing #AI #AIAgents #InfoQ

  14. has made its managed (MCP) server generally available, giving AI coding agents controlled access to AWS APIs, documentation & operational workflows through a standard interface.

    It offers a safer, more auditable way to connect AI agents to AWS services without exposing broad credentials.

    Learn more: bit.ly/49PyxL6

  15. Learn how to build a B2B intelligence platform with over one million company profiles using an MCP server - without creating an unsafe bridge between an LLM and production data.

    🔗 Read now: bit.ly/42EukGl

    #InfoQ #AI #AWS #AIAgents #ModelContextProtocol

  16. Learn how to build a B2B intelligence platform with over one million company profiles using an MCP server - without creating an unsafe bridge between an LLM and production data.

    🔗 Read now: bit.ly/42EukGl

  17. 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

  18. 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

  19. 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

  20. 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

  21. 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

  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. #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

  24. What happens when a major financial institution prepares its #APIs for #AIagents?

    #MorganStanley had to rethink large parts of its API program.

    The big shift: #ModelContextProtocol (MCP) - which has gone from obscurity to industry standard in ~18 months.

    It fundamentally changes who - or what - is consuming your APIs.

    Learn more: bit.ly/3PFnUmV

    #InfoQ