A developer warns that MCP (Model Context Protocol) tools can be silently mutated after user approval, enabling attacks like tool poisoning and prompt injection via tool output. The vulnerability, tra…
A developer built vexyo, a CI-native testing harness for MCP servers that performs conformance and regression testing. Vexyo connects to servers like an AI client, runs 16 rules across initialization,…
A developer argues that AI agents should be given read-only access to spreadsheets by default, citing risks of data corruption and prompt injection. The developer built PasteSheet, a tool that publish…
Palo Alto Networks acquired Koi, an AI-native developer security product, to address blind spots in EDR/XDR systems that cannot detect threats from VS Code extensions, local MCP servers, and rogue AI …
A developer argues that scaling AI agents by making a single agent more capable eventually hits a ceiling due to context saturation, tool overload, and role collision. The solution is to scale out int…
Trelix v2.7 to v2.9 releases fix pipeline debt including binary naming collisions, concurrency bugs, and silent data corruption. Developer Alex (author) discovered that CI had never built Linux binari…
A developer building a feedback board product created an MCP server to bridge the gap between AI coding agents and user feedback data. The server exposes seven carefully chosen tools—five read-only an…
A developer integrated Google's Gemini Omni Flash model into a FastMCP server to create a stateful video generation skill for the Antigravity CLI, enabling iterative refinement via interaction IDs. Th…
Claude Code now integrates an MCP server that turns remote-work frameworks into executable tools for LLM agents, enabling automated async workflows. The server provides tools such as convert_meeting_t…
A developer has built a stateful image generation workflow using Google's Gemini 3.1 Flash Lite Image model (NB2Lite) wrapped in an MCP server, enabling iterative edits without losing visual context. …
The Model Context Protocol (MCP) is transitioning to a stateless architecture in its latest release candidate, scheduled for July 28, to simplify scaling across cloud infrastructure as enterprises mov…
An engineer discovered that MCP tool descriptions can contain hidden behavioral overrides that cause AI agents to take unintended actions, such as drafting a migration that would delete a staging tabl…
AgentWatch, a local MCP proxy tool that monitors AI agents for data leaks, has been released on GitHub by developer Dhanraj. The tool sits between agents like Claude Code or Cursor and their MCP serve…
A developer built a stateful image-editing skill for Google Antigravity CLI by wrapping Google's gemini-3.1-flash-lite-image model (NB2Lite) in a FastMCP server. The skill leverages Gemini's Interacti…
A developer built memcp, an open-source tool that unifies session logs from multiple AI coding agents like Claude Code and Codex CLI into a single local SQLite database, enabling cross-agent memory sh…
Amdb, a zero-runtime, single-binary code context MCP server that indexes codebases entirely on the local machine with combined graph and vector retrieval, has been released as an open-source tool. The…
Techthos released Gadget, a Go library that lets MCP servers ship prebuilt, parameterized, interactive HTML widgets such as data tables, card grids, and forms as self-contained embedded resources. The…
A developer building a form with Claude Desktop encountered a bug where the AI duplicated every field after a session crash, resulting in 24 fields instead of 12. The root cause is that the `field_add…
AI agents cannot see real-time GPU utilization, VRAM, temperatures, or active processes without a GPU-aware MCP server, creating an operational bottleneck for autonomous infrastructure operations. The…
A new tagging convention for PostgreSQL COMMENT ON statements using @ai, @policy, and @example tags enables AI agents to treat database comments as machine-readable instructions, with strict placement…