Microsoft's Rayfin SDK brings typed data apps to Fabric
Rayfin is a new open-source TypeScript SDK for building Fabric Apps, working with humans via CLI and with AI agents, and turning data models into GraphQL endpoints with TypeScript clients.
Developers can define data structures and access controls in Rayfin, then let agents build on those models; docs include markdown files and a ready-to-run prompt for coding agents.
Source: InfoWorld
Rayfin is a new open-source TypeScript SDK for building Fabric Apps, working with humans via CLI and with AI agents, and turning data models into GraphQL endpoints with TypeScript clients.
Developers can define data structures and access controls in Rayfin, then let agents build on those models; docs include markdown files and a ready-to-run prompt for coding agents.
Source: InfoWorld
❤2
GitHub agentic autofix for code scanning alerts hits general availability
Assigning a code scanning alert to Copilot now explores your codebase, proposes a fix, reruns CodeQL to confirm the alert closes, then opens a draft pull request for review.
It replaces the free Generate Fix option with Assign to Copilot, needs an enterprise admin to enable the new Agentic autofix policy, and consumes AI Credits plus Actions minutes.
Source: GitHub Copilot
Assigning a code scanning alert to Copilot now explores your codebase, proposes a fix, reruns CodeQL to confirm the alert closes, then opens a draft pull request for review.
It replaces the free Generate Fix option with Assign to Copilot, needs an enterprise admin to enable the new Agentic autofix policy, and consumes AI Credits plus Actions minutes.
Source: GitHub Copilot
Cursor Cloud Agents Add Canvases, Environments API Beta
Cloud agents can now build interactive canvases as part of their work, shareable with your team by link. The Environments API is now available in beta.
Environments now support up to 300 repositories, and large environments load 5x faster. Use the beta API to programmatically create and manage the cloud environments your agents run in.
Source: Cursor
Cloud agents can now build interactive canvases as part of their work, shareable with your team by link. The Environments API is now available in beta.
Environments now support up to 300 repositories, and large environments load 5x faster. Use the beta API to programmatically create and manage the cloud environments your agents run in.
Source: Cursor
Pi Durable Brings Long-Running Agents to Cloudflare Durable Objects
Pi Durable is a harness for long-running agents. A new write-up builds a multiplayer chat agent with one Durable Object per session, hibernation, alarms, and subagents that search across conversations.
Sessions persist in Durable Object SQLite and wake for scheduled work, so agents need not run around the clock. A conformance suite checks custom storage; the demo needs Node.js 22.19+, pnpm 11, and an OpenAI API key.
Source: Earendil
Pi Durable is a harness for long-running agents. A new write-up builds a multiplayer chat agent with one Durable Object per session, hibernation, alarms, and subagents that search across conversations.
Sessions persist in Durable Object SQLite and wake for scheduled work, so agents need not run around the clock. A conformance suite checks custom storage; the demo needs Node.js 22.19+, pnpm 11, and an OpenAI API key.
Source: Earendil
❤2
Qoder CLI is now available
Qoder lists a command-line interface alongside its IDE and JetBrains plugin as another way to use the product.
Developers who prefer terminal-based workflows can now pick the CLI instead of the IDE or JetBrains plugin. The listing points to a download, so check the download page for setup details.
Source: Qoder
Qoder lists a command-line interface alongside its IDE and JetBrains plugin as another way to use the product.
Developers who prefer terminal-based workflows can now pick the CLI instead of the IDE or JetBrains plugin. The listing points to a download, so check the download page for setup details.
Source: Qoder
🔥2❤1
GitHub Copilot in Slack and Teams adds multi-repo support
Copilot in Slack and Microsoft Teams now works across multiple repositories in one conversation, so you can track pull requests, investigate issues, and review activity without switching conversations.
You can mention @GitHub anywhere in a message, switch repositories more easily, and let Copilot use shared Slack messages and threads as context. Reliability fixes target errors and task startup.
Source: GitHub Copilot
Copilot in Slack and Microsoft Teams now works across multiple repositories in one conversation, so you can track pull requests, investigate issues, and review activity without switching conversations.
You can mention @GitHub anywhere in a message, switch repositories more easily, and let Copilot use shared Slack messages and threads as context. Reliability fixes target errors and task startup.
Source: GitHub Copilot
❤7
Antigravity 2 turns the IDE into an agent chat app
Antigravity 2 drops the IDE for a chat-centric desktop app for agents. Workspaces are replaced by projects that span multiple directories with their own settings and permissions, and a new worktree mode gives each conversation its own Git worktree.
Projects replace single-directory workspaces and can span multiple directories with isolated settings and permissions. Worktree mode gives each conversation its own Git worktree, so parallel subagents don't conflict.
Source: InfoWorld
Antigravity 2 drops the IDE for a chat-centric desktop app for agents. Workspaces are replaced by projects that span multiple directories with their own settings and permissions, and a new worktree mode gives each conversation its own Git worktree.
Projects replace single-directory workspaces and can span multiple directories with isolated settings and permissions. Worktree mode gives each conversation its own Git worktree, so parallel subagents don't conflict.
Source: InfoWorld
❤4
Agentic Autofix now covers dependency and third-party alerts
Copilot Autofix for GitHub Advanced Security for Azure DevOps now runs on an agentic foundation, adds dependency vulnerability fixes, and supports code scanning alerts from third-party tools.
Enable it once at org, project, or repo level, then select alerts to request fixes. Review the diff and run builds and tests before merging; fixes consume AI credits and not every request produces one.
Source: Azure DevOps Blog
Copilot Autofix for GitHub Advanced Security for Azure DevOps now runs on an agentic foundation, adds dependency vulnerability fixes, and supports code scanning alerts from third-party tools.
Enable it once at org, project, or repo level, then select alerts to request fixes. Review the diff and run builds and tests before merging; fixes consume AI credits and not every request produces one.
Source: Azure DevOps Blog
❤3
Chrome DevTools 154 ships agent-focused JavaScript execution controls
Chrome 153 and Chrome 154 are now stable. DevTools adds fine-grained JavaScript execution controls for agents, AI source inspection and performance-trace recording, CPU performance-tier overrides, and full soft-navigation support.
If you build or evaluate AI coding agents, test the JavaScript execution controls first. Testing on low-end hardware? Try the CPU performance-tier overrides before third-party throttling extensions. The Application panel also gained ad-script tracking.
Source: shattered.io
Chrome 153 and Chrome 154 are now stable. DevTools adds fine-grained JavaScript execution controls for agents, AI source inspection and performance-trace recording, CPU performance-tier overrides, and full soft-navigation support.
If you build or evaluate AI coding agents, test the JavaScript execution controls first. Testing on low-end hardware? Try the CPU performance-tier overrides before third-party throttling extensions. The Application panel also gained ad-script tracking.
Source: shattered.io
❤6
mabl ships an agentic testing harness that checks coding agents' output
mabl's new harness gives teams independent verification of what coding agents build. It adds agent guardrails, a knowledge graph for impact and coverage gaps, terminal checks before a pull request, and test case sync with Xray, Zephyr, TestRail and Azure DevOps.
Teams using coding agents can define what good looks like and catch a change's impact before opening a pull request. Desktop application testing is coming soon for enterprises.
Source: SD Times
mabl's new harness gives teams independent verification of what coding agents build. It adds agent guardrails, a knowledge graph for impact and coverage gaps, terminal checks before a pull request, and test case sync with Xray, Zephyr, TestRail and Azure DevOps.
Teams using coding agents can define what good looks like and catch a change's impact before opening a pull request. Desktop application testing is coming soon for enterprises.
Source: SD Times
GitHub Copilot comes to Microsoft Teams via a consolidated app
The new GitHub app in Teams combines notifications with Copilot, so you can @mention GitHub on an issue notification and have it investigate, fix the bug, and open a pull request without leaving the chat.
If you have GitHub notifications installed, update to the latest version; existing notification features keep working. The legacy app becomes "GitHub (deprecated)" and retires on November 16, so install the consolidated app to keep using Copilot in Teams.
Source: GitHub Copilot
The new GitHub app in Teams combines notifications with Copilot, so you can @mention GitHub on an issue notification and have it investigate, fix the bug, and open a pull request without leaving the chat.
If you have GitHub notifications installed, update to the latest version; existing notification features keep working. The legacy app becomes "GitHub (deprecated)" and retires on November 16, so install the consolidated app to keep using Copilot in Teams.
Source: GitHub Copilot
❤1👍1
AWS Continuum is GA: AI agent opens fix PRs for code vulnerabilities
AWS made Continuum, its AI security agent for code vulnerabilities, generally available on Thursday. It finds and validates bugs in your repositories, checks them against your AWS environment, and opens pull requests with proposed fixes.
If you run repositories on AWS, Continuum can now open fix PRs against them, so expect agent-authored changes to review. The 89.0% figure comes from CyberGym-E2E, not your codebase; treat its PRs as proposals to verify.
Source: The New Stack
AWS made Continuum, its AI security agent for code vulnerabilities, generally available on Thursday. It finds and validates bugs in your repositories, checks them against your AWS environment, and opens pull requests with proposed fixes.
If you run repositories on AWS, Continuum can now open fix PRs against them, so expect agent-authored changes to review. The 89.0% figure comes from CyberGym-E2E, not your codebase; treat its PRs as proposals to verify.
Source: The New Stack
❤7
OpenAI's GPT-6.1 Terra Targets Visual Reasoning and Computer Use
GPT-6.1 Terra is priced at $1/million input tokens and $5/million output, almost half the price of GPT-5.6 Terra and GPT-6.1 Sol, and around 10x the price of GPT-6 Luna.
OpenAI emphasizes visual reasoning and computer use. Ted Sanders says Terra is better at technical communication, similar on vision and computer use but much faster, and a bit worse on most other things.
Source: Simon Willison
GPT-6.1 Terra is priced at $1/million input tokens and $5/million output, almost half the price of GPT-5.6 Terra and GPT-6.1 Sol, and around 10x the price of GPT-6 Luna.
OpenAI emphasizes visual reasoning and computer use. Ted Sanders says Terra is better at technical communication, similar on vision and computer use but much faster, and a bit worse on most other things.
Source: Simon Willison
❤2🤮1
Entire launches Git forge with Trails, an agentic replacement for pull requests
Entire, founded by former GitHub CEO Thomas Dohmke, unveiled a Git forge for hosting code, reviewing changes and running builds. Trails replaces the pull request with agentic code review.
Agent sessions sync into the repo for future use, and evals run on every push with findings in the web UI and open-source CLI. Private repos are natively hosted at launch; GitHub mirroring remains.
Source: GeekWire
Entire, founded by former GitHub CEO Thomas Dohmke, unveiled a Git forge for hosting code, reviewing changes and running builds. Trails replaces the pull request with agentic code review.
Agent sessions sync into the repo for future use, and evals run on every push with findings in the web UI and open-source CLI. Private repos are natively hosted at launch; GitHub mirroring remains.
Source: GeekWire
❤4
Junie Local swaps in Splash engine, hits ~70 tokens/s on M5 Max
Junie Local now delivers ~70 tokens/s on M5 Max for coding tasks, up from ~40 tokens/s, with bursts up to 107 tokens/s. Splash replaces MLX-VLM as its inference engine, and a DFlash2 draft model for Qwen3.8-3.6-27B-blend is included.
Faster decode and prefill change how usable a local coding agent feels on Apple silicon. Run junie update, then inside Junie run /local and press ctrl + r to update the inference engine. M4 Macs and newer with 64GB of RAM are now supported.
Source: JetBrains Junie
Junie Local now delivers ~70 tokens/s on M5 Max for coding tasks, up from ~40 tokens/s, with bursts up to 107 tokens/s. Splash replaces MLX-VLM as its inference engine, and a DFlash2 draft model for Qwen3.8-3.6-27B-blend is included.
Faster decode and prefill change how usable a local coding agent feels on Apple silicon. Run junie update, then inside Junie run /local and press ctrl + r to update the inference engine. M4 Macs and newer with 64GB of RAM are now supported.
Source: JetBrains Junie
❤1
GitHub Copilot Fusion is now generally available
Pick Copilot Fusion in the model picker instead of a single model, then choose Fusion Standard, High, or Max. It is available on all Copilot plans in VS Code, the Copilot app, Copilot CLI, and JetBrains IDEs.
Copilot routes and orchestrates models per task, so you stop switching models mid-task. Admins control which models Fusion can use and can set it as the default via Enterprise managed settings.
Source: GitHub Copilot
Pick Copilot Fusion in the model picker instead of a single model, then choose Fusion Standard, High, or Max. It is available on all Copilot plans in VS Code, the Copilot app, Copilot CLI, and JetBrains IDEs.
Copilot routes and orchestrates models per task, so you stop switching models mid-task. Admins control which models Fusion can use and can set it as the default via Enterprise managed settings.
Source: GitHub Copilot
Amp adds macOS Satellites: cloud Macs for iOS and Mac builds
Amp now has a Mac: ask it to work on your iOS or Mac app and watch it build, test, and run the app. Your thread still runs on a Linux orb; the Mac is a second machine attached to it.
Amp boots a fresh cloud Mac with Xcode and simulators in about 30 seconds, builds and tests there, and streams the screen so you can take control. macOS Satellites are in beta; join the waitlist in Settings → Experimental Features.
Source: Amp
Amp now has a Mac: ask it to work on your iOS or Mac app and watch it build, test, and run the app. Your thread still runs on a Linux orb; the Mac is a second machine attached to it.
Amp boots a fresh cloud Mac with Xcode and simulators in about 30 seconds, builds and tests there, and streams the screen so you can take control. macOS Satellites are in beta; join the waitlist in Settings → Experimental Features.
Source: Amp