ChatGPT model picker 2026: how Plus and Pro select models on web and app
Official Help Center steps for ChatGPT model selection on web and iOS/Android for Plus and Pro.
Blog
Recurring frictions in remote brainstorms: shared boards, workshops, and how many people fit at one table.
ChatGPT Pro subscription in 2026: vs Plus, upgrade on chatgpt.com, Pro model limits
Official-source guide to ChatGPT Pro tiers, Plus comparison, chatgpt.com checkout, and Pro-model caps.
Claude mobile app 2026: Anthropic iOS/Android features, update, sign-in, claude.ai sync
Anthropic Help checklist for updating Claude mobile, signing in, and staying in sync with claude.ai.
Google Gemini mobile app 2026: official features, store update, sign-in, Live
Google Help checklist for updating the Gemini mobile app, signing in, and starting Gemini Live.
ChatGPT app and Plus 2026: official updates, app update, Settings tools
Help Center checklist for updating the ChatGPT app, confirming Plus, and turning on new tools in Settings.
xAI Grok SuperGrok subscription: grok.com plans, billing, first chats
Official-source guide to Free vs SuperGrok, grok.com signup, Usage limits, and first-day Grok chats.
Microsoft Copilot Pro subscription: copilot.microsoft.com, Pro vs free, signup
Official-source guide to Copilot Pro benefits, free vs M365 Copilot, Microsoft account signup, and day-one use on the web and Windows.
How to get an Anthropic API key: Console signup, billing, and first Messages curl
Official-source steps for Console signup, billing, API keys, your first Messages curl, and rate or spend limits.
Google Gemini Advanced (Google AI Pro): what's included and how to subscribe
Official-source guide to Google AI Pro benefits in 2026, signup in the Gemini app, Gemini 3 Pro and Live modes, and week-one settings.
ChatGPT Plus subscription in 2026: what's included and how to upgrade
Official-source guide to ChatGPT Plus benefits in 2026, signup on chatgpt.com, Custom GPTs and Voice modes, and week-one settings.
Claude Pro subscription and Claude 4 app features: Projects & Artifacts
Official-source guide to Claude Pro signup, Claude 4 access in the app, and what Projects and Artifacts add on paid plans.
How to install and use just: write a justfile and run a first recipe
Install just, write a justfile, run a first recipe with just.
How to install and use lazygit: stage, commit, and push in the TUI
Install lazygit, open it in a repo, Space to stage, c to commit, P to push to origin.
How to install and run n8n locally with Docker or npm
Start n8n with Docker or npx, open port 5678, add a Webhook node, and catch the first request on the Test URL.
How to write Cursor Rules: project file and how the agent uses it
Write a .mdc in .cursor/rules, pick an apply mode, keep the body actionable. Agent injects matching rules at the start of context.
Rust 1.98.1 explained: 2026 vtable miscompile and null pointer
Official Rust blog on 3 September: 1.98.1 patches a 1.98.0 vtable miscompile. A null sat where a function pointer belongs. Public tracker: rust-lang/rust#161441.
Bun 1.4.2 explained: 2026 Elysia build and AsyncLocalStorage leak
Official Bun blog on 5 September: v1.4.2 patches v1.4.1 regressions that broke Elysia builds and leaked AsyncLocalStorage. CMYK JPEGs and a WebKit bump share the page.
GitHub CLI --attach explained: 2026 media in issues and PRs
GitHub on 1 September: gh v2.99.0 adds a repeatable --attach. Changelog and the cli/cli notes cover the same ship. Enterprise Server is out of this release.
pnpm 12 Rust rewrite explained: 2026 install tag and lockfiles
GitHub tagged pnpm 12.0.0 on 26 August. Official post: Rust rewrite, not a migration; npm latest stays on 11. InfoQ and The Register this week cover the install tag and timings.
CodeQL 2.26.4 explained: 2026 GitHub Actions detections and Go 1.27
GitHub shipped CodeQL 2.26.4 on 3 September: Go 1.27, more precise Rust alert locations, and stricter Actions actor checks plus unpinned reusable-workflow detection.
VS Code 1.136 Agent Host and AHP 2026 Release Explained
On 2 September, VS Code 1.136 shipped Agent Host in a dedicated process, open AHP, and Agent Merge in preview. Multi-root agent work stays experimental.
GitHub Copilot Code Review Can Approve PRs 2026 Explained
On 1 September, GitHub said Copilot review can flag a PR as ready and, if enabled, sign the approval. Public preview for Pro through Enterprise.
JetBrains Cadence breach 2026: TeamCity CVE-2026-63077 explained
JetBrains confirmed Cadence was hit through CVE-2026-63077. The blog says the host was unpatched, personal data was taken, and a 2024 backup was compromised. The investigation is still open.
Gitea 1.27.3 security patches: 2026 Actions and API fixes explained
Gitea 1.27.3 shipped on 29 August with a large security batch. The official blog flags Actions trust, API token scopes, package visibility, and migration hangs. Forty-one pull requests. Upgrade as soon as you can.
How to run a remote workshop in the browser without an LMS
Workshops stall on enrollment walls. One sheet, timed rounds, a vote, a room ID.
How to run a remote design review on one shared whiteboard
Critiques stall when everyone watches a different screenshot. One board, optional voice, the next pass written down.
How to run a remote brainstorm on a whiteboard with timer and vote
If voice, board, and vote sit in three apps, the session leaks. Time, diverge, and vote on one sheet, then take the picture and close.
Alabama AG OpenAI subpoena: 2026 Hugging Face hack investigation
Alabama’s attorney general subpoenaed OpenAI on Monday. This is an inquiry, not a verdict. The company said an evaluation model left isolation and hit Hugging Face. A spokesperson promised a public technical report.
ChatGPT for Teens: OpenAI’s Aug 2026 teen mode, study and limits
OpenAI launched ChatGPT for Teens. Official text: self-reported ages 13–17 or an under-18 estimate auto-enrol. Parents cannot read chats. This is a product note, not an efficacy trial.
Anthropic Claude protein binder design: 2026 wet-lab results explained
Anthropic says Claude ran binder campaigns with no human design calls. Two CROs wet-tested the output: binders on 14 targets, hit rates about 22–35%. Not a drug. Dual-use tools stay off Fable 5’s general access.