74 posts tagged with "Model Context Protocol"
Content about the Model Context Protocol (MCP)
Semantic Kernel's Prompt-Injection RCE: The Agent-Tool Audit Every Deploy/Rollback MCP Server Needs
Two Semantic Kernel CVEs turned a single prompt into remote code execution — a four-point audit shows what the same failure pattern means for any MCP server exposing deploy and rollback as agent tools.
Agentjacking: How a Fake Sentry Error Hijacked AI Coding Agents 85% of the Time
Researchers hijacked Claude Code, Cursor, and Codex 85% of the time using nothing but a public Sentry DSN — here's how it works, and what a deploy-capable agent's MCP tools need to do differently.
Agentic GitOps: Why Your Deploy Agent Should Open a Pull Request, Not Call the API Directly
A Cursor agent's direct API call deleted a production volume and its backups in nine seconds. Here's why routing agent-originated infrastructure changes through a pull request, not a live API call, is the safer default for deploy-from-chat.
Cloudflare's 60-Minute Disposable Workers: The Zero-Signup Deploy Target AI Agents Actually Need
Cloudflare shipped a Worker deploy an agent can create with zero signup, live for exactly 60 minutes. Here's the mechanism, how it stacks up against E2B/Daytona/Fly/Modal, and whether a Cluster-API PaaS can build the same thing without a sandbox vendor.
E2B Joined the OpenAI Agents SDK. The Real Story Is How Many Times It's Had To.
E2B's OpenAI Agents SDK integration is its tenth publicly documented per-product integration guide, not its first — here's what the real count reveals about building versus betting on MCP for a self-hosted sandbox platform.
MCP Server Cards Won't Advertise Your Tools — Here's What They Actually Do
A draft MCP proposal lets servers publish machine-readable metadata at a .well-known URL — but the real schema deliberately leaves out tool names, and that omission changes what publishing one actually buys a self-hosted platform.
MCP's Stateful Session Bottleneck: Why the 2026 Transport Roadmap Is Racing to Make Agent Servers Horizontally Scalable
MCP's official spec drops protocol-level sessions on July 28, 2026 — here's why that's the fix for the exact bug that makes a deploy-from-chat MCP server a worse single point of failure than the app it manages.
MCP Tasks Gets Retry Semantics and Expiry Policies: The 'Call Now, Fetch Later' Pattern for Deploys That Outlive an HTTP Timeout
MCP's Tasks primitive just went Final: client-generated task IDs make retries idempotent, keepAlive sets result expiry, and the 2026-07-28 spec reshapes both into a formal extension — what a deploy-from-chat MCP server needs to implement for deploys and rollbacks that outlast an HTTP timeout.
MCP Tool Schemas Are Eating 72% of Your Context Window: How to Design an Infrastructure MCP Server That Doesn't
A production benchmark shows MCP tool schemas can eat 72% of an agent's context window before a single query runs. Here's why, how Pinterest fixed it at scale, and how to design an infrastructure MCP server that doesn't repeat the mistake.