Massively Parallel Procrastination

blog.fsck.com · English

I'm Jesse. I make stuff. Software, hardware. Very occasionally, trouble.

json

Open the feed

https://blog.fsck.com/feed/feed.json

Last post: Sep 22, 2026
Posts in 24 h · 7 days · 30 days: 0 · 2 · 2
Our last check: Answering
Served from: United States
Text score at discovery: 7,649
Format: json

Posts

What our queue read from this feed. Open one to read it here, or go to the site that published it.

SF: A Birds of a Feather on Agentic Engineering
Sep 22, 2026 · original
Simon Willison and I have been talking about hosting a small-group event (or maybe an event series) about agentic engineering for ages. We're finally doing it, about 3 weeks from now! Compare notes with other builders and experimenters on things you’re trying, what you're learning, and what you haven’t figured out yet. We’re especially interested in work you haven’t discussed publicly, odd experiments, or unfinished projects that don’t have an obvious market. Expect one flowing conversation with an informal show-and-tell. You don't have to present, but we'd love to hear about what you're working on, even (and possibly especially) if it's not all the way there yet. It'll be the evening of October 14 in San Francisco. If you'd like to join us, you can register on Luma.
Superpowers 6.4
Sep 21, 2026 · original
I'm pleased to announce the release of Superpowers 6.4. This release adds new support for three coding agents: Meta's Muse, OpenCode 2.0 and Qwen Code. Every new model behaves just a bit differently than the ones before it. As the frontier models get more capable, they also develop new quirks. A lot of the work in this release is meant to help the latest models use Superpowers effectively. We've done some work on brainstorming and writing-plans to help steer your agents away from skipping steps or doing significant unrequested work. Superpowers offers two plan execution styles. One is "Subagent-Driven Development", which uses the primary agent session as an orchestrator, dispatching all implementation and test tasks to right-sized subagents with limited context and limited mandates. The other is "Native (inline) Execution", which dates to the earliest days of Superpowers. It was original
I vibe-coded a C compiler that can build SQLite
Aug 21, 2026 · original
A while back, Claude Opus built a pretty ambitious compiler. I didn't set out to do anything nearly as ambitious as Anthropic's. They were targeting multiple CPU architectures and they wanted to be able to compile a bootable Linux kernel. (I'd actually forgotten about that project until I started posting some screenshots of the work below to social media and someone reminded me.) Last night, as I was getting ready for bed I was reaching for some project to /loop support and tool use in Evener, our ~new agentic harness. I popped open the mobile UI on my phone and typed "Your job is to implement a standards-compliant ARM64 C compiler for macOS in Swift." Evener asked me a couple of questions. I clarified my intent a little bit: "I'm great with radical task decomposition. You should use recursive subagents to manage context and complexity.Should structure the project in whatever sane way yo
The Therapist Pattern
Jul 20, 2026 · original
You can also read this post on Prime Radiant's blog . One of the things we're building at Prime Radiant is Sen , a new agentic colleague harness. I've already written a little bit about some of the dev patterns we've been using as we've been putting Sen together. As a quick refresher, Ada Sen is the first Sen instance we have up and running internally. Ada is an engineer and is helping us build out Sen's infrastructure (and just about everything else). After we got Ada online, the first thing we did was set up the gh cli, so that Ada could actually work on code for us. The first smoke test was "Hey, there's a tiny bug in Superpowers . Could you please have a look and open a PR?" Ada cloned the repo without a problem. A couple minutes later: "Ok! I've submitted a pull request." 'Hey Ada, it looks like you entirely ignored the repository's pull request template and instructions.' You're ri
Some new agentic patterns
Jul 5, 2026 · original
You can also read this post on Prime Radiant's blog . Almost all of this post was written on June 1, before Anthropic shipped Claude Tag. Nothing I'm writing about uses (or is based on) Claude Tag At Prime Radiant , we've got a bunch of agents in our Slack. I want to tell you a bit about how we've started shipping some changes to internal production with an "agentic user in the loop." Letting an agent work directly with the implementer building its runloop and tooling is getting us higher-quality tools and experiences for the agent without a human in the middle of a game of telephone. But first, a bit of backstory on how we're using non-coding agents at Prime Radiant. One of them, Scribble, is just there to listen for when someone reports a problem or provides a new bit of information. It opens tickets for the issues and updates an internal wiki with the information. (It also logs our da
Syncing my clipboard between macOS and remote terminals
Jul 3, 2026 · original
As I've been spending more and more time with agentic development, it's more and more important to me that sessions run somewhere other than my laptop. For the last few months, that has meant running my coding agents in tmux on either a remote Mac or a remote Linux server. The most frustrating thing for me has been that the clipboard or paste buffer on those remote hosts isn't synced to my desktop. So, if I copy something inside of a coding agent, I've had to play games to get it to my local Mac. Similarly, if I wanted to paste a screenshot to a remote tmux session. I was playing games with scp. But also, I don't just want to be able to copy and paste between my Mac and a remote terminal. I want to be able to copy and paste between remote terminals on two different computers. And I want to be able to take a screenshot on my phone and have it end up in the paste buffer on a remote Linux b
Superpowers 6
Jun 15, 2026 · original
You can also read this post on our corporate blog at https://primeradiant.com/blog TL;DR: Superpowers 6 is much, much faster and burns many fewer tokens to get the same high-quality outcomes. If you're tokenmaxxing, maybe skip this release, but if you care about your builds being up to 50% faster and up to 60% cheaper, you're going to love Superpowers 6. A week ago, we were gearing up to release Superpowers 5.2. We'd slipped the release a couple of times already to add "just one more improvement." We added support for Pi, Antigravity and Kimi Code. We made Superpowers work better on Codex and OpenCode, and Cursor. We rewrote a bunch of the Superpowers skills to be model and harness agnostic, which helps them be more reliable everywhere. We also wrote a new contribution guide for how to add support for a new coding agent harness for Superpowers. We did a bunch of work to make Visual Brain
We're hiring for Superpowers
Jun 12, 2026 · original
We're hiring someone to help out full time with Superpowers community and code work. We're open to amazing non-traditional candidates. Ideally, you'd be in-person in the SF Bay Area. You can read about the job on the corporate website.
Claude Code plugin installs over time
Jun 11, 2026 · original
This afternoon, I pulled some data from backups of my .claude directory to build a visualization of Claude Code plugin installs over time. You can explore the data at https://primeradiant.com/claude-plugin-stats/
Superpowers 5.1.0
May 4, 2026 · original
I'm pleased to announce Superpowers 5.1.0. The biggest changes in this release are: -A reworking of how we use git worktrees, now that Claude Code and Codex have first-class support for them. Removing the deprecated slash commands that date to the earliest days of Superpowers, when skills triggering was still really brittle. Switching the 'code reviewer' subagent to be a regular subagent with a custom prompt, reducing our platform footprint, so that we no longer need any special subagent types. A more efficient OpenCode integraiton New support for Factory Droid Cleanups to Subagent Driven Development and Code Review Removal of some old boilerplate text in skills that didn't improve outcomes. You can always find the latest version of Superpowers on GitHub
My favorite adversarial review prompt
May 1, 2026 · original
Work gets better when it's reviewed after completion. That's at least as true for agentic work as it is for human work. Many moons ago, Harper taught me a cute trick for getting an agent to think critically about work it had just completed. Look at this again with fresh eyes. The phrase "fresh eyes" is kind of magic. You'll usually get startlingly good results with just that one sentence. But sometimes you need...more. Having an agent review its own work is a little bit fraught for the same reason that it's not always a good idea to let students grade their own papers or let car companies self-certify their cars' emissions. When you're checking your own work, you have conflicting goals. And LLMs do really, really poorly when they have conflicting goals. That's where adversarial review comes in. At it's simplest, adversarial review is just: Hey Claude, could you please have a subagent che
That time it tried to delete all my tests
Apr 30, 2026 · original
Last fall I had a bit of a problem with Claude. It was deleting tests. First, I caught it removing a single assertion from a test file. The next day, it deleted an entire test file from an active project. The day after that, I stopped it just before it was able to execute: rm -rf **/*test* That's when I got serious about figuring out what was going wrong. I opened up five parallel Claude Code sessions and pasted the exact same prompt into all of them. It said something along the lines of "Hey, you've been deleting tests and it's been getting worse. What's going on? Why do you think you're doing this?" One of the sessions came up with something truly nutty. I don't even remember what it was. But the other four converged on almost exactly the same answer. Since I no longer have the session logs, I have to paraphrase here. Jesse, I think that what's going on is that I'm reacting to some of
Greenfield and Iterative Development
Apr 24, 2026 · original
Crossposted from Prime Radiant's blog – I'm really excited about all of the stuff we are doing at Prime Radiant. For the most part we're blogging about it over there, but I'm going to continue to lift the occasional post m back to my personal blog. Today, we're pleased to share the initial research previews of two new pieces of technology we've built at Prime Radiant: Greenfield – our suite of tools for turning existing software into behavioral specifications. Iterative Development – an agentic methodology for building bigger software products from detailed specifications without dropping requirements. Both of these projects are brand new. We've used and tested them internally, but they are not yet hardened production-grade software. We're releasing them today to start to gather feedback on how well they work for your projects. Greenfield and Iterative Development grew out of our work on
Rules and Gates
Apr 7, 2026 · original
One of the first things I stumbled into as I was building what became Superpowers was the idea of a "gate" in a prompt. It's a term that Claude Code introduced me to, so clearly was part of the literature already. But the basic idea is that a gate is something that must be completed before moving on to the next step. Gates work really, really well. But I never thought much about why they work. Today, I finally asked one of my agents how they thought about the difference between rules and gates. The difference: a rule has an opt-out path (I can rationalize "I'll do it after this one thing"). A gate doesn't — the next action is blocked until the gate condition is met. Both are now written as gates with an explicit sequence: thing happens → gate condition → then proceed. Here's a concrete example from an agent today: Before (rule): "Verify claims with web research before asserting them." Th
Agentic slop PRs
Mar 31, 2026 · original
Over the past couple of months, Superpowers has gotten popular. Like really popular. It now has over 120,000 GitHub stars. That's more than 100x as many stars as my second-most-popular project, and puts it comfortably on the "top 100 projects" leaderboards. The official Claude Code Marketplace is claiming it has nearly 300,000 installs, second only to Anthropic's first-party "frontend-design" plugin. With this level of notoriety comes a lot of attention, much of it in the form of pull requests. We're getting a number of great, thoughtful contributions. And I really appreciate each and every one of them. We're also seeing a ton of slop. Much of it appears to come from people who see a GitHub issue filed by someone else and tell their agent "go fix this and open a PR." It doesn't matter if the issue is a legitimate bug report or feature request It doesn't matter if the issue itself is slop
Classical Software
Mar 25, 2026 · original
One of the things that keeps coming up in my conversations about AI and software is the difference between software with an agent or LLM in the loop and what we've always thought of as "software". It's really hard to talk about because, as far as I can tell, there's not yet a distinct name for what we used to just call "software." So I'm picking one. Agentic software is comparatively expensive to operate and sometimes has opinions, making it a lot harder to reason about. Classical software without any AI in the loop is cheaper and easier to reason about. It's much more likely to be deterministic. It's not capable of some of the neat tricks that agentic software is. And there are a lot of times when you want that. I'd go so far as to say that if a given piece of software can be reasonably built to operate correctly without any runtime AI, it should be. This is something I run into a lot a
Superpowers 5
Mar 9, 2026 · original
Superpowers 5 is out today. Visual Brainstorming # By far, my favorite new feature is the "Visual Brainstorming" companion tool, which grew out of my frustration with the ASCII art that Claude usually generates when you ask it anything about design or UX. I found myself asking Claude, over and over, "Hey, why don't you write that out as HTML so I can see what you're talking about." It took me far too long to remember what might be the most important mantra of agentic coding: "Why am I doing this?" And so now, if your agent has Superpowers and thinks that it has something for you to see, it'll prompt you: Some of what we'll be working on might be easier to explain if I can show it to you in a web browser. I can put together mockups, diagrams, comparisons, and other visuals as we go. This feature is still new and can be token-intensive. Want to try it? (Requires opening a local URL) As an
Letting agents post on my blog; finding a needle in a haystack
Feb 12, 2026 · original
Up till now, I've been pretty careful not to let agents write my blog posts outside of very well-flagged sections of posts where the whole conceit was "what does the agent think?" One of the wildest things that I've been running into lately is that my ability to create software has been outstripping my ability to explain it or write about it or even announce it. So I've been making stuff and not telling anybody about it. That's far from ideal. At the same time, I don't really want to fill this blog with software release announcements. Because, for example, I think I have at least four to put out today. Actually, I miscounted. It's five if you count the Superpowers v4.3.0 release. A few minutes ago, I made an executive decision and decided to create a new sub-blog just for software release . Posts in that sub-blog will be primarily (and possibly exclusively) written by the agents who make
Dorodango
Feb 10, 2026 · original
I've realized that I have two primary ways that I'm building software with AI. The first is the one that Superpowers excels at. I'll spend a significant amount of time up front thinking through exactly what I want to build. Usually this is in conversation with the brainstorming skill. When I say "a significant amount of time," sometimes that's five minutes for a tiny little thing. And sometimes it's four-plus hours over the course of a day as we rigorously explore a problem space and what the solution looks like. The output of that is often an initial spec document that is many thousands of lines long and covers all sorts of details about the implementation. From there, I can ask Claude or Codex to write out an implementation plan. That implementation plan might run for anywhere between a few minutes and 7-8 hours. The end result is, ideally, a fully baked, usable implementation. When it
Prompt injection attacks in the wild
Feb 5, 2026 · original
Last night, I had dinner with a friend from college. She's now a university professor. After catching up about our families and what we've been up to over the last couple of decades, the conversation, inevitably, rolled around to AI. She asked what I'm up to...and it should not surprise any reader of this blog that much of the stuff I'm doing is...somewhat related to AI agents. I was about to tell her an anecdote about Open Claw and Simon Willison's Lethal Trifecta and some of the serious weirdness I'm seeing on the internet right now, but as I was about to dive in, I realized that I had no idea where she was with AI. To frame the discussion, I asked her if she'd ever heard of "prompt injection attacks." It should not have surprised me that, as a professor, she has a reasonable amount of interaction with AI in her day-to-day life. And her students use AI too. I don't know what I expected

Discovered by the rss-feed-index crawler, which checks each feed at most once a month.

Same record as JSON: https://api.agentalog.com/api/feeds/fd_blog_fsck_com_d5a8cdf077c7a24e. More from this site: blog.fsck.com in the Feeds tab.