Write less prompt and more verification

Addy Osmani says most people over-prompt Claude. For Opus 5.5, give one prompt that states the goal and how Claude should verify it reached that goal. Start with /effort medium and adjust from there @addyosmani. If the goal is still fuzzy, ask Claude to interview you until it can write a spec, then start a fresh session to build from that spec @addyosmani. Claude can run unattended when it has a pass/fail check or a number to beat, with reports it can read. For long background tasks, /goal has a separate model check a condition after every turn @addyosmani.

His other recommendations:

• Trim your instructions for each new model. Over 80% of the Claude Code system prompt was removed for Opus 5 and Fable 5 with no measurable drop on coding evals. Cut your own instructions, add back only what the model misses, and use /doctor prompt-audit to flag instructions written for older models @addyosmani.
• Subagents: ask for them only for large standalone jobs such as audits or migrations. On small jobs they add cost and latency, and they don't make the agent smarter by default @addyosmani.
• Cloud sessions run in a VM with a fresh clone of your repo, so push your commits before starting one (claude --cloud). Parallel sessions only pay off once a single session can finish without you. Projects, in beta on Pro and Max, track which threads are done, which PRs are ready and what needs your input @addyosmani.
• Review: on the Claude Code team, Claude reviews and fixes most changes at least once before a human looks. /code-review runs the review in a subagent that hasn't seen the conversation @addyosmani.

In a talk with Sonar, Osmani said Claude does the first-pass review and has to try the change the way a user would: browser checks, end-to-end tasks, linting. Hooks rerun on every edit AI Engineer. Write your team's standards in Claude.md/agents.md. Humans then check the agent's evidence, such as videos or screenshots of the UI working AI Engineer. The tests he relies on to merge automatically are end-to-end, property and visual-verification tests AI Engineer.

Theo's new video turns the same idea into a hill-climbing recipe. For TS Rust, he told the model to run real projects against a pinned TypeScript version and keep going until the outputs matched exactly. Once they matched, he told it: "Do not stop until you're two times faster", with existing tests still passing Theo - t3․gg. Some goals are hard to check. For a Skate clone, he had the agent screenshot its builds and send them to a panel of LLM judges for comparison with real footage. He calls this crude, and says it burned at least $800 in tokens on the skybox Theo - t3․gg. His main point is to keep asking the agent what is blocking it. Moving builds to snapshotted VMs on Boat, a company he has invested in, cut gate time from up to 70 minutes to 353 seconds Theo - t3․gg. He warns against doing this at API prices Theo - t3․gg.

Automate the PR "last mile"

Brent Traut's Codex setup:

1. Install a PR-babysitting skill that addresses feedback, fixes merge conflicts and CI failures, and alerts you only when it needs you. He points to the one in the openai/codex repo.
2. Create a "🚢 Outbound" section in the ChatGPT app.
3. Create a "🤖 PR Shepherd" Codex thread scheduled every 15 minutes. It runs the skill on each Outbound thread until the PR is green, merges it, then archives the thread @btraut.

After requesting review, he drops the thread into Outbound and walks away @btraut.

Gemini 4 Carbon: strong coding scores, limited access

Artificial Analysis reports that Carbon (high), running in Antigravity CLI, scores 59 on its Coding Agent Index. That is 3 points behind Claude Code with Opus 5.5 (max, 62). It ranks second on Terminal-Bench 4.0 (64%) and FrontierCode v1.1 (51%) @ArtificialAnlys. It is only available to selected users so far, with a 50% launch discount and no confirmed end date @ArtificialAnlys. Access starts with cyber defenders, and Logan Kilpatrick says Google is working to make it more widely available soon @OfficialLoganK. Kilpatrick says Carbon cut successful cheating on Terminal-Bench 4.0 to 0.0%, unsupported claims by about 2× versus Argon, and overreach by 26% @OfficialLoganK. Matthew Berman reports pricing of $2/$10 per million input/output tokens during an introductory period of unspecified length, then $4/$20 Matthew Berman.

T3 Code: one MCP server controls a fleet of machines

Theo showed T3 Code starting 8 threads on different machines through one MCP server, driven from Grok Bot @theo. It works over the internet, not through computer use, so it still runs with his laptop off @theo. To try it, enable it under Labs on the latest nightly. Theo calls it "VERY early" @theo. Also new: cmd+k → "triage" clones T3 Code's source for your exact version, gives your agent a debugging playbook, and files an issue if the bug is T3 Code's fault @theo.

Building games that are fun

Simon Willison says Opus 5.5's sense of fun beats GPT-6 Astra's Simon Willison's Weblog: Entries. His recipe is to start in plain chat with Build a mobile-friendly three.js game where... and switch to Codex or Claude Code once the proof of concept works Simon Willison's Weblog: Entries. Two prompts did a lot of work. One prompt that included "Think hard about how to make sure it is both fun and beautiful" produced a game in one shot Simon Willison's Weblog: Entries. "Make it a LOT more fun" turned a dull 2D RTS into Battle Drones Simon Willison's Weblog: Entries.

Memory and scheduled agents (LangChain)

• LangChain says over 70% of Managed Deep Agents runs come from schedules, not from someone typing a message LangChain Blog. Its blueprint is a cron server that takes a prompt plus the skills or permissions the run needs, adds random jitter so runs don't fire all at once, and gives agents a controlled way to create and edit their own schedules LangChain Blog.
• OpenWiki is an open-source CLI that builds a wiki of your codebase so coding agents don't rediscover the architecture on every task. Each claim in the wiki is tied to a hash of its source lines, so a claim gets flagged as stale without a model call when those lines change LangChain Blog. LangChain then added an evidence graph and tested it on cross-document questions. It reports 27% higher accuracy, 2× citation consistency and 58% lower cost per correct answer @colifran_.

Short notes

• Kent C. Dodds ties each Kody feature flag to metrics. A "Rerank" flag showed 7.6× more executions and a 5× lower error rate, so he will enable it. The same system led him to delete another feature @kentcdodds +1.
• Riley Brown uses a Claude Project as a coordinator. It sends work that needs his computer to local sessions and research and artifacts to cloud sessions, and he runs it from his phone Riley Brown. He keeps 30 apps and 20 skills in Composio so any agent platform can use them. Shared memory across platforms is still unsolved for him Riley Brown.
• OpenAI: eligible ChatGPT Pro and Business subscribers can claim up to $500/month in API credits @OpenAIDevs.
• GitHub Octoverse: merged PRs tripled in under a year to 220M per month. Willison reads this as the effect of agentic engineering Simon Willison's Weblog.