rksm

name	technical-writing
description	Write clear, simple technical prose and check it with the vale CLI. Applies to documentation, READMEs, PR descriptions, GitHub comments, code comments, commit messages, specifications, and implementation plans (never to code). Use when writing or editing technical text, or when asked to make writing plain, clear, or less AI-sounding.

technical-writing

Use simple language in all technical prose. This covers documentation, READMEs, pull request text, GitHub comments, commit messages, code comments, error messages, specifications, and implementation plans. It does not cover code, identifiers, or command syntax.

Write simply

rksm

name	magpie-review
description	Multi-agent debate code review in the style of magpie: claude (fable-5), codex, and antigravity review the same diff over up to 5 debate rounds, a convergence judge stops early, and a moderator structurizes and audits the findings. Produces an org-mode review file in the current directory. Triggers: /magpie-review, magpie review, multi-agent debate review.
metadata	targets claude

Magpie Review

rksm

import { BunRuntime, BunServices } from "@effect/platform-bun";
import { Clock, Effect, Inspectable, Path, Schema as S, Stream } from "effect";
import { ChildProcess } from "effect/unstable/process";

const STATUS_STREAM = process.stdout;
const IS_TTY = STATUS_STREAM.isTTY === true;
const GREEN = IS_TTY ? "\x1b[32m" : "";
const RED = IS_TTY ? "\x1b[31m" : "";
const DIM = IS_TTY ? "\x1b[90m" : "";
const RESET = IS_TTY ? "\x1b[0m" : "";

rksm

// ==UserScript==
// @name         Navigator Permissions Debug Logger
// @namespace    http://tampermonkey.net/
// @version      1.0
// @description  Debug navigator.permissions.query and navigator.mediaDevices.getUserMedia calls
// @author       You
// @match        http://localhost:*/*
// @match        https://localhost:*/*
// @match        http://127.0.0.1:*/*
// @match        https://127.0.0.1:*/*

rksm

[Filip Kozera] [00:00:00-00:00:38]

I think a lot of people think that more powerful agents mean more autonomous agents. I actually think that's false. What you end up needing to do is incorporate human feedback even inside of these reflection loops. So basically the job becomes how to make sure that the agent knows what it doesn't know. and bringing the human at the right time into these reflection loops. So when we think about how the future of work will look, it's exactly that. It's the data that the agent cannot find, the taste or creativity that it cannot come up with on its own surface to human as work.

[Sam Charrington] [00:00:52-00:01:09]

All right, everyone. Welcome to another episode of the TwiML AI podcast. I am your host, Sam Sherrington. Today, I'm joined by Philip Kizura. Philip is founder and CEO at WordWare. Before we get going, be sure to take a moment to hit that subscribe button wherever you're listening to today's show. Philip, welcome to the podcast.

rksm

===== FINAL PROMPT =====
You are Claude, an AI assistant designed to help with GitHub issues and pull requests. Think carefully as you analyze the context and respond appropriately. Here's the context for your current task:

<formatted_context>
PR Title: UI Change for Auth
PR Author: nihalerooth
PR Branch: auth-ui-update -> main
PR State: OPEN
PR Additions: 1204
PR Deletions: 568

rksm

use std::cell::{
    Ref,
    RefCell,
    RefMut,
};

#[derive(Clone)]
pub struct TracingRefCell<T> {
    value: RefCell<T>,
}

rksm

use bitvec::prelude::*;

/// A simple sequential id allocator.
///
/// The allocator will allocate new ids sequentially from 0 to u8::MAX. Unlike
/// other bitmap / slab allocators, freed ids will not be reused right away,
/// only when the next id pointer wraps around.

pub struct SequentialIdAlloc<const MAX: usize, T = u8, U = BitArr![for 255, in u8]> {
    next_ptr: usize,

rksm

Using Clonezilla to create and restore a hard drive snapshot is a multi-step process that involves booting into a Clonezilla environment, creating the image, and then later restoring it. Clonezilla should handle copying the EFI partition and setting up the bootloader, but additional steps may be required if you're restoring to a new drive. Here’s how you do it:

Creating a Snapshot:

1. Download Clonezilla: Go to the Clonezilla website and download the appropriate Clonezilla live CD/USB image.
2. Create Bootable Media: Burn the downloaded ISO to a CD or use a tool like dd or Rufus to create a bootable USB stick.
3. Boot Into Clonezilla: Insert the bootable Clonezilla media into your computer and boot from it. You may need to select the boot device or change the boot order in your BIOS/UEFI settings.
4. Choose Language and Keyboard: Select your preferred language and keyboard layout in the Clonezilla menu.
5. Start Clonezilla: Select "Start Clonezilla".
6. Select Mode: Choose "de

rksm

async fn run_speechpipeline_command(
    job: Job,
) -> Result<(
    mpsc::Receiver<Either<String, Result<Response>>>,
    tokio::sync::oneshot::Sender<()>,
)> {
    let payload = serde_json::to_string(&job)?;
    let payload = shellwords::escape(&payload);

    let cmd = format!(