# Opus 5.5 Clears a Stalled Port; the Assembly Speedup Has a Narrow Benchmark *By Coding Agents Alpha Tracker • September 26, 2026* Two practitioners report Opus 5.5 completing major language translations, while the linked `ttfx` PR gives a testable, decrypt-specific speed result rather than a general coding benchmark. The brief pairs that signal with concrete practices for scoping long runs, verifying agents, and evaluating new developer tools. ## 🔥 TOP SIGNAL **Long-horizon code translation is the strongest signal.** Theo says Opus 5.5 got his four-month, on-and-off TypeScript-compiler-to-Rust port working in 10 hours, after GPT-5.6 Sol and GPT-6 Astra plateaued at roughly 35% and 85% of tests passing. [^1][^2] DHH separately says Opus 5.5 translated Omarchy’s `ttfx` from Rust to x86-64 assembly and claims up to 17× faster. The linked PR reports 14.5× on one `decrypt` test (371→25.5 ms) and separately lists 32.5 ms versus Rust’s 371 ms, with 206/206 oracle cases matching—strong project-specific evidence, not a general coding benchmark. [^3][^4] ## ⚡ TRY THIS - **Give long runs a finish line and a quality target.** Theo’s exact request: `/goal finish the port and make it faster`. Peter Steinberger used `/goal` with Astra to move OpenClaw off synchronous SQLite after it became limiting with one agent running 50 sessions in parallel; he reports 575 PRs landed so far, with improvements shipped along the way. [^2][^5] - **Make review adversarial and isolated.** Steinberger’s auto-review skill starts a fresh-context reviewer, feeds its findings back to the writer, and repeats; then the agent gets a separate test box to install the app, inspect the UI, and verify it end to end. He says this can turn a 10-minute PR into two hours, but lets him parallelize about 20 tasks—reserve the extra validation for changes where it pays back. [^6] - **Turn Slack bug reports into repro-to-PR jobs.** Boris Cherny says Claude Tag writes more than 50% of his PRs each day; his prompt is: “@Claude from now on, attempt to repro every bug in this channel end to end, running the full app. Once you reproduce the bug, put up a PR to fix it and tag the right team for code review.” [^7] [Claude Tag](https://claude.com/product/tag)’s personal connectors are rolling out on Team plans, with Enterprise to follow; those personal connectors are for user-requested work, not scheduled or agent-started runs, which require connectors attached to the channel. [^8] - **Write the HLDD after the implementation works.** Geoffrey Huntley’s pattern: use pre-commit hooks and agent skills to keep the high-level design doc updated, then self-host it in the product. Put the human-authored rationale and success measures first, implementation details below, and editor remarks where a subsection needs its intent preserved. [^9] ## 📡 WHAT SHIPPED - **Antigravity 2.0 adds `/plan` in the app and CLI.** The workflow explores files and dependencies without writing code, asks clarifying questions, then produces a reviewable plan with verification checkpoints. Review or adjust it and click **Proceed**; the docs say whether the agent pauses for approval is configurable through the artifact review policy, so set that explicitly if you need a hard gate. [^10][^11] - **Microsoft Copilot Code is an app builder, not the new default coding IDE.** It turns natural-language requests into dashboards, automations, and internal apps in a sandbox that can be hosted in a tenant. Frontier rollout is slated for month-end, Managed Runtime is already in preview, and Microsoft says developers will keep using GitHub Copilot for day-to-day work. [^12] - **Autopilot is a separate persistent agent.** Microsoft describes a cloud-hosted agent that watches channels, follows up, handles recurring work, and can resume projects days later; its blog says private-preview expansion is at month-end. Omar Shahine says it is built on OpenClaw, whose team worked with Microsoft on deployment readiness; local inference, file transfer, and code mode on machines connected to the OpenClaw gateway are still described as upcoming. [^12][^13][^14][^15] - **Cua Driver is now stable and open source for Omarchy.** TryCua says its native synthetic cursor for Hyprland enables OS-level multi-cursor control. The [repo](https://github.com/trycua/cua) shows two sessions operating Calc and Inkscape on Omarchy and supports agent connections through CLI, MCP, or typed SDKs. [^16][^17] - **OpenRouter introduced `typesafe/jev-router`, a cache-aware model-and-reasoning-effort selector.** Theo’s objection is coding-specific: Jev categorizes but cannot see the codebase or the agent’s tools, so prompt-only complexity estimates may miss the work involved. He also doubts mid-task downgrades save much when cache writes are a large cost, and says he is still gathering numbers; his early comparison to a “DeepSeek 4.1 Flash router” is only an impression. [^18][^19][^20][^21] ## 🎬 GO DEEPER - **Video — [Peter Steinberger on coding-agent review loops](https://www.youtube.com/watch?v=zrGXl7m76lY):** The useful segment is his fresh-context reviewer and isolated test-box workflow, not a model demo. [^6] [![Peter Steinberger: Stop Sending Me Code. Send Me Your Prompt](https://img.youtube.com/vi/zrGXl7m76lY/hqdefault.jpg)](https://youtube.com/watch?v=zrGXl7m76lY&t=2040) *Peter Steinberger: Stop Sending Me Code. Send Me Your Prompt (34:00)* - **PR — [Omarchy `ttfx` Rust-to-assembly port](https://github.com/omacom/ttfx/pull/35/):** Inspect how Rust remains the front end and fallback for unported effects, while the assembly engine is checked against byte-for-byte oracle cases. [^4] - **Repo — [Cua Driver](https://github.com/trycua/cua):** The quickstart connects an agent to Calculator, asks it to compute 6 × 7, and checks that the app displays 42; the repo also documents CLI, MCP, and SDK integrations. [^17] **Editorial take:** The edge is not simply “let it run”: specify the finish line, then give an independent reviewer and a real execution environment a chance to falsify the result. [^2][^6] --- ### Sources [^1]: [𝕏 post by @theo](https://x.com/theo/status/2103665601466237261) [^2]: [𝕏 post by @theo](https://x.com/theo/status/2103558185487577476) [^3]: [𝕏 post by @dhh](https://x.com/dhh/status/2103595410921279635) [^4]: [Add an x86-64 assembly engine for all 37 effects by dhh · Pull Request #35](https://github.com/omacom/ttfx/pull/35/) [^5]: [𝕏 post by @steipete](https://x.com/steipete/status/2103648679169257737) [^6]: [Peter Steinberger: Stop Sending Me Code. Send Me Your Prompt](https://www.youtube.com/watch?v=zrGXl7m76lY) [^7]: [𝕏 post by @bcherny](https://x.com/bcherny/status/2103538666597691552) [^8]: [Claude Tag now supports personal connectors in channels](https://claude.com/blog/claude-tag-now-supports-personal-connectors-in-channels) [^9]: [𝕏 post by @GeoffreyHuntley](https://x.com/GeoffreyHuntley/status/2103421056765853984) [^10]: [𝕏 post by @antigravity](https://x.com/antigravity/status/2103611698800140697) [^11]: [Plan \(/plan\)](https://antigravity.google/docs/plan) [^12]: [Introducing the new Copilot with Home, Code and Autopilot - The Official Microsoft Blog](https://blogs.microsoft.com/blog/2026/09/25/introducing-the-new-copilot-with-home-code-and-autopilot/) [^13]: [𝕏 post by @OmarShahine](https://x.com/OmarShahine/status/2103480227561079264) [^14]: [𝕏 post by @steipete](https://x.com/steipete/status/2103491173927272531) [^15]: [𝕏 post by @steipete](https://x.com/steipete/status/2103491221515907087) [^16]: [𝕏 post by @trycua](https://x.com/trycua/status/2103498682532253734) [^17]: [trycua/cua: Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.](https://github.com/trycua/cua) [^18]: [𝕏 post by @OpenRouter](https://x.com/OpenRouter/status/2103610898690855161) [^19]: [𝕏 post by @theo](https://x.com/theo/status/2103614549006156146) [^20]: [𝕏 post by @theo](https://x.com/theo/status/2103614897129144545) [^21]: [𝕏 post by @theo](https://x.com/theo/status/2103701465084919906)