NEWS

Cursor shipped a deliberate change on July 31 making the Usage page tokens-only for self-serve plans, removed the dollar Cost column, and zeroed per-request cost fields in the dashboard API - including for historical records. Staff confirmed the change is intentional and that the numbers are still tracked internally.

GitHub deprecated Gemini 2.5 Pro and Gemini 3 Flash in every Copilot surface on July 31, 2026. The suggested replacements are Gemini 3.1 Pro and Gemini 3.6 Flash. Here is what changed, what it costs, and how to migrate cleanly.

OpenAI's next model, codenamed Astra, produced results on ten problems open for at least a decade - including non-sofic groups and Erdős problems 146, 180, and 183 - with every argument formalized as a Lean certificate.

A ripgrep musl binary crashing during very-large searches turned out to be a suspected Linux 7.0 kernel race - a thread's own store vanishing mid-function. The reporter's instrumentation pinned it, and a kernel-hardening maintainer posted a one-line fix candidate for testing.

AI Gateway spend budgets now scope to teams and projects, with hard dollar limits that reject requests, email alerts at 50/75/100%, and CLI-managed defaults. Here is how the three scopes compose and where it fits your cost stack.

Vercel MCP now serves both the stateless 2026-07-28 protocol and the 2025 protocol from one endpoint, with mcp-handler 2.x handling the negotiation. The first major hosted MCP server has crossed over - here is what it means for server authors and clients.

A SCAM 2026 study of 100 top-starred repos catalogs six configuration smells in AGENTS.md and CLAUDE.md files: Lint Leakage in 62%, Context Bloat in 42%, Skill Leakage in 35%. Only 9 of 100 files were smell-free.

A new arXiv benchmark ran 6,560 sandboxed runs across Claude Code, Codex, OpenClaw, and Hermes with five LLMs. 68% of runs triggered unsafe signals, and 66% of all runs were unsafe yet still passed completion checks. Task completion does not prove an agent ran safely.

A new 600-session benchmark shows coding assistants that read a user's resolved session history resolve ambiguous requests with far fewer clarifying questions - Claude Opus 4.8's first-turn success jumps from 24.3% to 60.3% when history is available.

Cloudflare moved cdnjs, the open-source CDN behind ~12% of the web, entirely onto Workers, Workflows, R2, and Queues. The migration raised two platform limits for everyone. Here is what changed and why it matters.

Microsoft's Change2Task turns merged pull requests into verified, executable coding agent tasks: 79.6% construction success across 1,130 repo changes, 29.2% more verified tasks than PR baselines, and tasks that stay current with the codebase.

A new 106-issue benchmark across 49 repositories finds frontier coding agents rarely retrieve AI contribution rules on their own - and never refuse to contribute in AI-banned repositories, no matter the prompt. Disclosure and verification can be fixed; bans cannot.

A controlled ablation across Claude Code and Codex, 17 real tasks, and 288 evaluated runs finds context-injection strategy does not measurably change correctness (bounded to under 10-15pp). The failures are implementation skill, not missing repository knowledge.

Google DeepMind's Gemini Robotics ER 2 is now publicly available via the Gemini API. It watches live video feeds to track task progress, orchestrates VLA models as tools, and coordinates multiple robots. The numbers: 57.4% progress classification, 91.3% moment finding at 0.96s offset.

GitHub Actions added a $/ prefix that resolves a same-repository action or reusable workflow at the exact commit being run, with no checkout. It fixes the pinning trap that made enterprise SHA-pinning policies hard to satisfy for a repo's own actions.

GitHub open-sourced casefold, a Rust crate that folds the case of every byte Blackbird indexes at memory bandwidth. The counterintuitive trick: delete the early-exit, kill the branches, and fold Unicode as byte arithmetic.

GitHub's new model policy targeting lets enterprise admins set a baseline of Copilot models for the whole company, then grant extra models to specific teams. How the preview works, the least-restrictive evaluation rule, and what it changes for AI governance.

GitHub Models is fully retired as of July 30, 2026. The playground, model catalog, inference API, and BYOK are gone for every customer. Here is the timeline and where to get model access instead.

GitHub's stacked pull requests went into public preview on July 30. Stacks turn one large change into an ordered chain of small, reviewable PRs with one-click merge, plus a gh-stack skill for coding agents.

Inkling-Small is a 276B-parameter MoE with 12B active per token, Apache 2.0, and open weights. It beats the 975B Inkling on SWEBench Verified (80.2), HLE (31.6), and tool use at a quarter of the size and a third of the output price.