MCP - Page 12
Stories from MCP come first, newest first, then stories from connected topics (AI Agents, AI Agents, prompt injection, Developer Tools, agent harness, coding agents), each labelled with where it comes from. This view is not indexed by search engines.
Latest
METR and Redwood Say They Found Gaps in OpenAI’s Agent Monitoring
Researchers from METR and Redwood spent three weeks inside OpenAI testing its agent monitoring and found several flaws. OpenAI says they are patched.
Google Cloud Unveils Agent Compute Engine to Run AI Agents in MicroVMs
Google Cloud's Agent Compute Engine gives each AI agent its own hardware-isolated microVM. It reaches preview in the coming months; no GA date was given.
Anthropic’s New Claude API Version Rejects Temperature Settings
Anthropic's Claude API version 2026-10-21 returns 400 errors for temperature, top_p and top_k. New SDKs send it by default and move to Apache 2.0.
GitLab Fixes Second 10.0-Rated Regex Code-Execution Flaw in 10 Days
GitLab has patched CVE-2026-110264, a CVSS 10.0 flaw letting unauthenticated users run code. Versions back to 7.0 are affected; self-managed must upgrade.
Cognizant Launches AI Nexus, a Five-Part Layer to Meter and Route AI
Cognizant has launched AI Nexus, five modules to track AI spend, coach quality and route work across models. It gave no pricing or named customers.
NVIDIA Releases Five Open Nemotron 3.5 Retrieval Models for AI Agents
NVIDIA's open Nemotron 3.5 Embed, Rerank and Parse models find, rank and extract evidence for AI agents. The benchmarks are NVIDIA's own.
Anthropic’s Claude Duet Agent Will Not Move Money Even With Approval
Anthropic has detailed Claude Duet's safeguards: approval cards, a two-tier privacy classifier, passwords the model never sees, and actions always barred.
Grok Bot Can Now Build Reports, Memos and Dashboards as Pages
Grok Bot's official X account says it can now build reports, memos and dashboards that users edit with it or have it keep up to date. Few details given.
Recursive Releases Flint, an Open-Weight Coding Model Built to Run Locally
Recursive has released Flint, a 40B open-weight coding model that runs locally. It says AI built most of it; all benchmark and speed figures are its own.
Blacksmith Says Sequoia Round Values Its Code-Testing Cloud at $1bn
Blacksmith, which runs builds and tests for software teams, has raised a $66m Series C led by Sequoia at a $1bn valuation, two months after its Series B.
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.