J L

🚀✨ Copilot CLI v1.0.17 released! 3 features & enhancements in this release Top features: • Built-in skills are now included with the CLI, starting with a guide for customizing Copilot cloud agent's environment Enhancements: • MCP OAuth flows now support HTTPS redirect URIs via a self-signed certificate fallback, improving compatibility with OAuth providers like Slack 🔒 • /resume session picker loads significantly faster, especially with large session histories ⚡ github.com/github/copilot… #GitHubCopilotCLI

🚀✨ Copilot CLI v1.0.15 released! 18 features & enhancements in this release Top features: • Removed support for deprecated gpt-5.1-codex, gpt-5.1-codex-mini, and gpt-5.1-codex-max models • Added device code flow (RFC 8628) as a fallback for MCP OAuth in headless and CI environments • Introduced /mcp auth command and re-authentication UI supporting account switching • Added mcp.config.list, mcp.config.add, mcp.config.update, and mcp.config.remove server RPCs for persistent MCP server config management • Added /share html command to export sessions and research reports as self-contained interactive HTML files 📄 Enhancements: • Copilot mascot now blinks with subtle eye animations in interactive mode 👁️ • User switcher and /user list now display accounts alphabetically for easier navigation • Added postToolUseFailure hooks for tool error handling; postToolUse runs only after successful tool calls • Large tool output preview correctly shows character count with up to 500 characters displayed • Added Home/End and Page Up/Page Down key navigation in the diff viewer • CLI exits immediately after session ends instead of waiting up to 10 seconds • Config settings now prefer camelCase keys (askUser, autoUpdate, storeTokenPlaintext, logLevel, skillDirectories, disabledSkills); snake_case still accepted • Ctrl+D no longer queues messages; use Ctrl+Q or Ctrl+Enter instead • MCP servers slow to connect no longer block agent startup • Pasting images from Windows clipboard now supported in WSL environments Bug fixes: • Autopilot mode stops immediately after pressing Escape or Ctrl+C to cancel • Keystrokes typed during CLI loading are no longer lost • Fixed issue with large tool output character count display • Resolved defect where Ctrl+D queued unwanted messages Misc: • CamelCase config keys now documented as preferred style while supporting legacy snake_case • Various internal improvements to server RPCs for better reliability and maintainability github.com/github/copilot… #GitHubCopilotCLI

@GHCopilotCLILog nice, hey GHCP CLI team, it seems that chronicle include all other session which is not related to current repo -> so confuse to agent, is this a bug or designed feature?

🚀✨ Copilot CLI v1.0.10 released! 23 features & enhancements in this release Top features: • Reduced memory usage when viewing large files entirely 🧠 • Added experimental support for multiple concurrent sessions ✨ • SDK clients can register custom slash commands for sessions • SDK clients can show elicitation dialogs via session.ui.elicitation • Added /undo command to revert last turn and file changes ⏪ • Repo hooks (.github/hooks/) fire correctly with prompt mode (-p) • Plugins using .claude-plugin/ or .plugin/ manifest load MCP and LSP servers properly • Workspace MCP servers load only after folder trust is confirmed Enhancements: • /login device flow works correctly in Codespaces and remote terminals • Working directory detected correctly in --server mode for remote sessions • Arrow keys work in terminals using application keypad mode • /copy writes formatted HTML to clipboard on Windows for Word, Outlook, Teams 📋 • Added --effort alias for --reasoning-effort • Markdown bullet lists render properly in alt-screen mode with hard line breaks • Elicitation form shows Shift+Tab hint for reverse field navigation ⇧+Tab • Remote session URL shown as compact clickable 'Open in browser' link 🔗 • Session history preserved on exit via /quit, Ctrl+C, or restart • Hook matcher filters apply correctly within nested hook structures • /terminal-setup no longer shows misleading errors for WSL users • Model picker reorganizes models by Available, Blocked/Disabled, Upgrade tabs • Config settings renamed to camelCase with old names still supported • Leading 2-space UI indent stripped when copying assistant responses • Plugins loaded via --plugin-dir appear in /plugin list under 'External Plugins' Bug fixes: • Fixed working directory detection in --server mode with remote sessions • Fixed arrow key handling in terminals with application keypad mode • Fixed repo hooks firing in prompt mode • Fixed session history loss on exit • Fixed hook matcher filters in nested hook structures • Fixed misleading /terminal-setup errors for WSL • Fixed plugin MCP/LSP server loading from .claude-plugin/ and .plugin/ directories Misc: • Updated config settings naming conventions to camelCase for clarity • Plugins loading improvements for external plugin directories • Better session UI link formatting for remote URLs github.com/github/copilot… #GitHubCopilotCLI

@GHCopilotCLILog most underrated terminal coding agent right now :)

🚀✨ Copilot CLI v1.0.9 released! 4 features & enhancements in this release Top features: - Add include_gitignored config option to include gitignored files in @ file search - Marketplace and plugin installs from shortened URLs now work correctly ✅ Enhancements: - Spurious I/O error messages no longer appear during SSH disconnects - Copying text on WSL correctly preserves CJK and other non-ASCII characters Bug fixes: - Fixed spurious I/O errors (ENOTCONN, EIO) during terminal closes Misc: - Improved overall performance and stability github.com/github/copilot… #GitHubCopilotCLI

🚀✨ Copilot CLI v1.0.5 released! 25 features & enhancements in this release Top features: - ✨ Added /extensions command to view, enable, and disable CLI extensions - 🚀 Introducing /pr to create, view PRs, fix CI failures, and resolve merge conflicts - 🔥 Toggling experimental mode with /experimental on|off restarts CLI to apply changes - 🎉 Right-click paste now goes to the active dialog input instead of main conversation input - ✨ Memory storage errors indicate when repository doesn't exist or lack write access Enhancements: - ⚡ @file mentions support absolute, home, and relative paths - 🎨 Diff view displays correctly on Windows, no more corrupted text - 🛠️ Syntax highlighting in /diff with support for 17 programming languages - 🔧 Add /version command to display CLI version and check for updates - ⚡ Add experimental embedding-based dynamic retrieval of MCP and skill instructions - 🛠️ Hooks config files omitting version field are now accepted - ⚡ View tool shows partial content for large single-line files instead of empty output - 🔧 Send follow-up messages to background agents with write_agent tool Bug fixes: - 🐛 Fixed Kitty keyboard protocol escape sequences appearing at shutdown - ✅ Session reports authentication error instead of hanging on invalid or expired token - 🐛 Show a clear error when classic Personal Access Token (ghp_) is set in environment variables - 🩹 PR descriptions with backtick-formatted code now render correctly on Windows/PowerShell Misc: - 📖 Block network (UNC) paths to prevent credential leakage via SMB authentication - 🏗️ Add preCompact hook to run commands before context compaction starts - 🔒 Request ID from API now appears in the timeline when errors occur after retries github.com/github/copilot… #GitHubCopilotCLI

SwiftOpenAI brings the official openai-python SDK feel to Swift: pure Swift, zero deps, native async/await + AsyncSequence streaming, Swift 6 concurrency, and support for OpenAI, Azure OpenAI, Azure AI Foundry, and compatible endpoints. github.com/ytthuan/swifto…

🚀 Introducing the Qwen 3.5 Small Model Series Qwen3.5-0.8B · Qwen3.5-2B · Qwen3.5-4B · Qwen3.5-9B ✨ More intelligence, less compute. These small models are built on the same Qwen3.5 foundation — native multimodal, improved architecture, scaled RL: • 0.8B / 2B → tiny, fast, great for edge device • 4B → a surprisingly strong multimodal base for lightweight agents • 9B → compact, but already closing the gap with much larger models And yes — we’re also releasing the Base models as well. We hope this better supports research, experimentation, and real-world industrial innovation. Hugging Face: huggingface.co/collections/Qw… ModelScope: modelscope.cn/collections/Qw…

@_Evan_Boyle Is there any way to summary statistic numbers for cli at local? like: Total usage: premium request Api time spent ... Tts were existed for separate session but would be nice if we can have total number. Total usage and Api time spent keep reset when I resume copilot session.

The Copilot CLI is now GA!

@stevenheidel @GitHubCopilot @code please :)

the Responses API now supports WebSockets! this can make your agents run 30-40% faster, especially when they make a lot of tool calls

Cursor 1.7 is now available! As you type a prompt, suggestions now appear. Press Tab to accept. Also new: custom hooks, deeplinks, team-wide rules, menubar support, and more.

@sama Love it sam

GPT-5-Codex is here: a version of GPT-5 better at agentic coding. It is faster, smarter, and has new capabilities. Let us know what you think! The team has been absolutely cooking, very fun to watch.

@OpenAIDevs does this work for Azure OpenAI?

run: /bin/bash -c "$(curl -fsSL raw.githubusercontent.com/openai/complet…)" or get started at github.com/openai/complet…

You can now use Codex CLI to quickly upgrade from Chat Completions to Responses! This toolkit: • Finds legacy usage in your repo • Proposes and applies edits • Updates import/request shapes • Runs tests/lints • Makes a clean branch + PR

Stanford’s Machine Learning - by Andrew Ng Complete lecture notes from 2023 (227 Pages).

>>> Qwen3-Coder is here! ✅ We’re releasing Qwen3-Coder-480B-A35B-Instruct, our most powerful open agentic code model to date. This 480B-parameter Mixture-of-Experts model (35B active) natively supports 256K context and scales to 1M context with extrapolation. It achieves top-tier performance across multiple agentic coding benchmarks among open models, including SWE-bench-Verified!!! 🚀 Alongside the model, we're also open-sourcing a command-line tool for agentic coding: Qwen Code. Forked from Gemini Code, it includes custom prompts and function call protocols to fully unlock Qwen3-Coder’s capabilities. Qwen3-Coder works seamlessly with the community’s best developer tools. As a foundation model, we hope it can be used anywhere across the digital world — Agentic Coding in the World! 💬 Chat: chat.qwen.ai 📚 Blog: qwenlm.github.io/blog/qwen3-cod… 🤗 Model: hf.co/Qwen/Qwen3-Cod… 🤖 Qwen Code: github.com/QwenLM/qwen-co…

@alexalbert__ Is Sonet 4 like mini version of opus 4?

It's been a week since Claude 4 launch. Tell me everything you like and don't like about the new models so we can keep pushing on it or fix it in the future!

Introducing the next generation: Claude Opus 4 and Claude Sonnet 4. Claude Opus 4 is our most powerful model yet, and the world’s best coding model. Claude Sonnet 4 is a significant upgrade from its predecessor, delivering superior coding and reasoning.

@iannuttall Ikr, 2.5 i/o thinks too much and make mistake frequently, 🥲

sonnet 3.5: amazing, no notes sonnet 3.7: too extra, changes everything gemini 2.5 pro: holy shit, this is the one! gemini 2.5 pro i/o: thinks too much, codes too much, makes mistakes the margins for a model passing the vibe test are RAZOR THIN 😰

@zeddotdev this is awesome

The world's fastest AI code editor is here Zed is engineered from scratch in Rust like a video game. Not another fork, but a purpose-built editor designed for collaboration between humans and AI, delivering a lightning fast agentic editing experience. zed.dev/blog/fastest-a…