I see incredible power in @cursor_ai Automations. Right now, they seem totally unusable (for my use case). I want to integrate Cursor Automations with Linear and trigger an automation when an Issue moves to 'Todo' status. The only way I can figure out how to set up the Cursor Automation trigger is when 'Status Changes.' This means that a new agent spawns every time the status changes, not just when the status = 'Todo.' So when Agent A moves an issue from Todo to Done, Agent B (that has no reason to exist) is spun up. Agent A, as its final step, moves the next logical issue to Todo, triggering Agent C, and so on. So we end up with an Agent waterfall... Without being able to get more specific on triggers, I have to either use a Cron trigger (definitely don't want to do that) or I really just can't use Automations. Any ideas? @leerob

.@leerob Trying to embrace Cursor Automations, but there's a friction point for me: I'd love some help drafting the agent instructions to ensure I don't miss anything. If I could explain what I want the automation to do and have a model engineer more robust instructions, I'd have launched several already. What do you think?

.@mattpocockuk I’m taking over someone else’s codebase and need a quick way to understand the code. You got a skill for that? 🙃😆

Using Opus 4.7 after working with Composer 2.5.

Insanely well-deserved. I just keep getting more and more value out of these. Just used /prototype for the first time and blew my mind. 🤯 Thank you, @mattpocockuk !!!

This is so true! I used to have multiple things I’d be working on with Opus, but with Composer, by the time I switch to the other task I was going to work on, it’s already responded on the fist!

I was using Claude Code for the last several months. I felt like I was getting way more bang for my buck. The Composer 2.5 results, however, were extremely intriguing. So I put $20 down on @cursor_ai just to test Composer 2.5, which is exactly as advertised and more. What I didn't expect, however, was that while I was away, Cursor team made the DX absolutely elite! I am loving the (new?) Agents window. It's so fast. The hotkeys are exactly what they should be. And Composer 2.5 is slaying it. I'm back. Well done, Cursor team.

Built a plan in Claude with Opus 4.7 and @mattpocockuk's /grill-with-docs skill, then gave it to @cursor_ai's new Composer 2.5 model (which executed absurdly fast), then went back to Claude and asked Opus 4.7 to audit all uncommitted changes against our intent: "The work is genuinely high-quality. Typecheck and lint are clean, conventions are mostly respected, and the architecture matches ADRs 0007/0008/0009 closely." Bullish on the future of xAI and Cursor...

Absolutely loving the speed of Composer 2.5 so far. But the 200K token context window is a little jarring after working with 1M Opus 4.7 for the last several weeks.

Claude or Codex and why?

I am in love with the /improve-codebase-architecture skill, but I found myself wanting to take on more than is wise at once. So I created docs/architecture-backlog.md and edited the skill to consider this document before its exploration phase as a subtle clue to what we've already considered in the past, why it wasn't the right time to make that change, and weigh whether now is a better time given the new state of the repo. @mattpocockuk good idea? terrible idea? Does my concern make sense and would you do it differently?

For the last few days, I have been in the habit of shipping at least one PR using this skill from @mattpocockuk before I’ve finished my morning coffee. It has completely transformed my repo. And whatever I build the rest of the day benefits disproportionately from this work.

The /improve-codebase-architecture from @mattpocockuk is 🔥 I'm using AI heavily to build my app, and I've worked hard to keep it DRY and clean throughout the build. Regardless, DRY has been violated many times, and some bloat has occurred. I've run this skill just once, and it has found so much that can be optimized for the longevity of the code base!

New favorite use for Claude Desktop: using the @posthog connection, have claude create all of my dashboards, insights, etc. 🤌🏼

It's April 2026 and Opus 4.7 still estimates 1-2 weeks of work for this task in our todos. 4 min later and we've shipped to prod.

One of the simplest tactics I've started using to 10x my workflow with Claude: "Any pushback? Challenge this UI/UX decision."

Are you kidding me? You can interrupt a sleep to move to the next step in a @vercel @WorkflowDevKit run?! What an amazing DX for testing and development.

Shipping with docs bundled is 🔥🔥🔥