Coding

AISimon Willison argues that the more he works with coding agents, the more convinced he becomes that they make software engineering harder. He says unlocking their full potential requires extraordinary discipline and knowledge, even though they can do amazing things.

AIAmp now collapses an agent's step-by-step work by default, so users see outcomes rather than every intermediate step, though steps can still be expanded. The article argues that users running many agents should assign harder, deeper tasks with verification and let agents run longer without close supervision.

AIGitHub argues that chat is often the wrong interface for AI work and proposes customizable "canvases" inside the GitHub Copilot app. Canvases are full-stack applications running without browser chrome that can communicate bi-directionally with the Copilot agent and execute code locally. The post cites examples including a Connect 4 game, a Winget package manager UI, and a SQLite database interface.

AIHugging Face released SmolDataEnvs, more than 5,000 open-source RL environments aimed at real-world data science tasks. They target the gap between simple educational games and frontier-level benchmarks, especially for improving coding in models under 10B parameters. The environments are designed as a testbed for developing new RL methods such as GRPO or OPSD.

AIGitHub Security Lab describes the Fuzzing Taskflow, an LLM agent pipeline that identifies entrypoints, writes harnesses, runs AFL++, reads coverage reports, and triages crashes for C/C++ repositories. The agent makes decisions while MCP tools handle execution, and state is stored in a SQLite database. The post also warns that the taskflow runs AFL and build commands directly on the host, so it should be used only in disposable environments without elevated privileges.

Why it matters: The post explains how an LLM agent automates fuzzing steps like harness writing, coverage gap chasing, and crash triage, with a runnable workflow and design tradeoffs.

AIKilo (@kilocode) announces that agent swarms are now available and asks developers how they plan to use them in development. The post gives no details on features, pricing, or availability.

AIAmp users can now share a runner with their workspace by starting it with --share, letting everyone spawn threads on that machine from ampcode.com. Shared runners appear under Shared Runners in the picker, and --amp-env gives them workspace and project Secrets & Env Vars but never personal ones. Amp warns that collaborators run code as the owner with their files and credentials, so sharing should be limited to trusted people, and workspace admins can disable runner sharing in Member Settings.

AIThe Amp macOS app now starts a runner automatically, so threads can run on your Mac without keeping amp --no-tui open in a terminal. Users add folders or projects under Runner in App Settings, then select "This Mac" when starting threads from ampcode.com, a phone, or Puck. A Keep This Mac Awake option prevents sleep while the runner is on and the Mac is plugged in, though the screen still turns off and locks.

AIClaude builds a model of a program's most complex parts, such as state machines or race-prone code, and searches that model for counterexamples that signal suspected bugs. It then reproduces those bugs and fixes them in the code. The post clarifies that the whole codebase is not formally verified, only the riskiest sections are modeled and checked.

AIClaude Code cloud sessions are now generally available, letting work continue while a laptop is closed. Existing Pro subscribers get a one-time $100 credit and Max subscribers get $250 to try them.

AICursor has made Rollouts and Security Reviewer available today on Teams and Enterprise plans. The company is including Rollouts usage credits for the next 10 days so customers can try it on real changes. More details are available in Cursor's blog post.

AICursor's Security Reviewer now completes reviews in 3.8 minutes on average, down from 4.8 minutes, a 21% speedup. The update was released alongside Rollouts.

AIAnthropic published a write-up on how it made the Claude frontend faster using agentic optimization techniques. The post coincides with Max Woolf's separate blog post showing that prompting agents can make code faster than current state-of-the-art libraries, with prompts and benchmark results included.

AIFelix Rieseberg, who works on Claude at Anthropic, says he feels joy when people use it to create things he did not think possible. The quoted post from @The_Alex reports early access to Opus 5.5 and shares several creations, including a Dark Souls project.

AICursor announced improvements to the token efficiency of its agent harness, linking to a blog post with the details. The post itself offers no figures or specifics beyond that headline claim.

AIMinecraft creator Notch says on X that he is enjoying vibe coding and admits he may have been slightly wrong. Months earlier he had publicly rejected AI-written code, but he later began having AI build internal tools such as a map editor and node graph tools. The main post adds that he has accumulated a set of small tools for himself before much game development has happened.

AIMike Knoop argues that formal verification is becoming feasible and is important for security. He adds that it does not automatically build human understanding, which he calls an even bigger alignment problem. The post is framed as a reply to Boris Cherny's report that Claude Opus 5.5 helped formally verify the Claude Agent SDK in Lean, producing 16 bug-fix PRs.

AIAnthropic's Cat Wu points users to ways to try Opus 5.5 and Claude Tag in Slack. The main post carries no further details on access, features, or availability.

AICursor is launching two bots, Rollouts and Security Review, for Teams and Enterprise plans. Rollouts monitors each pull request as it deploys and reports change health per environment, while Security Review reports exploitable bugs on every pull request.

AISwitching effort levels mid-session on Opus 5.5 does not break the prompt cache, according to Lydia Hallie of Anthropic. This holds when running Claude Code v2.1.280 or later.

AIKilo Code tested Grok 4.7 against Opus 5.5 on a touching-grass simulator, with Grok's one-shot result costing 47% less. Per the quoted post, Grok 4.7 cost $3.52 versus $7.35 for Opus 5.5, and both models are available in Kilo now.

AIA prompt shared by Alex Albert asks Claude to recreate San Francisco's Market Street as it stood on April 17, 1906, before the earthquake, using Blender. It requires building a source file from Sanborn fire insurance maps, the Miles Brothers film, period photos, and USGS topography, with reusable Blender Python generators for facades, street lamps, and vehicles, ending in a 10-second video up the street.

AIAnthropic has launched Claude Opus 5.5, the first model in its new Claude 5.5 family. According to the quoted launch post, it performs at the level of Claude Fable 5.1 for most tasks and costs 40% less to run than Opus 5. The author says it leads in coding and knowledge work and praises its writing quality.

Why it matters: The quoted launch post gives a concrete cost comparison, useful for weighing Opus 5.5 against earlier Opus and Fable 5.1 models for routine work.

AIAnthropic introduced Claude Opus 5.5 as the first model in its Claude 5.5 family, saying it performs at the level of Claude Fable 5.1 for most tasks at 40% lower run cost than Opus 5. Boris Cherny reports that Opus 5.5 and Fable 5.1 each ported HAProxy from C to Rust and both passed nearly all of its tests, with Opus 5.5 finishing in 9.5 hours versus 12 hours and at 51% less cost.

Why it matters: The author reports a same-task comparison in which Opus 5.5 finished a HAProxy C-to-Rust port faster and cheaper than Fable 5.1, offering a concrete cost and time benchmark.

AIClaude Opus 5.5 is now the default model in Claude Code and the Claude app, including Cowork, for Pro, Max, and Team plans. Anthropic is defaulting to effort medium across products, which it says is comparable to Fable 5.1 on intelligence but faster. Rate limits will go 25% further on Opus 5.5 compared to Opus 5.

Why it matters: The post names concrete default changes across Claude Code and the Claude app, plus a specific effort setting and rate-limit difference, useful for judging day-to-day cost and speed.

AILovable now offers Opus 5.5, which it says matches Opus 5's results while finishing builds in a third to half fewer steps. Internal benchmarks showed Opus 5.5 scoring 4 to 6% ahead of Opus 5 on verification discipline, with step reductions of 26% to 57% and input token reductions of 21% to 59% across tasks.

AIStepFun released an install script for Step Code that runs on macOS, Linux, or WSL with a single curl command. WSL is the recommended route on Windows, while PowerShell support is currently in beta. The project accepts issues and pull requests on GitHub.

AITencent Hunyuan introduced WebCraftBench, a benchmark that tests AI agents by using the live web app and scoring aesthetics, usability, and whether the original request was met. Coverage-guided exploration reaches parts of the app that agents otherwise miss. On 197 human-validated pairs, the benchmark matches human preference 85.3% of the time.

AIAmp runners can now create Git worktrees from the directory picker, with each new worktree placed as a sibling folder on a new branch off the current HEAD while uncommitted changes stay in the original checkout. Runners can also opt in with --amp-env to inject Secrets & Env Vars configured on ampcode.com into thread shell commands, MCP servers, and plugins, with changes applied to the next thread without a restart.