Skip to content

Instantly share code, notes, and snippets.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Pellentesque interdum augue libero. Nullam vehicula purus a dui lobortis, eu aliquet nunc efficitur. Nullam ultricies mauris sit amet libero elementum pellentesque. Integer odio nisl, sagittis vel neque vel, tincidunt elementum tellus. In at erat lectus. Suspendisse ut dignissim nulla, a gravida leo. Vivamus ullamcorper convallis porttitor.
Proin sodales commodo massa eu iaculis. Mauris accumsan faucibus urna quis euismod. Aliquam erat volutpat. Cras semper felis felis, feugiat dapibus erat malesuada sed. Aenean tristique diam non leo mattis, vel molestie odio ornare. Donec eget tincidunt lectus. Aliquam erat volutpat. Donec dictum tortor non pellentesque placerat. Sed consequat enim sit amet pellentesque dictum. Sed non lobortis neque. Nullam imperdiet eu urna vitae feugiat. Ut eu molestie ante, sit amet pulvinar metus. Aenean id orci est. Nunc arcu est, convallis rutrum finibus vel, dignissim a nunc.
Vivamus sit amet elit eget nibh placerat efficitu
#!/bin/bash
# Script to download, unpack, and run the Buildkite MCP server
# Usage: ./scripts/run-buildkite-mcp-server.sh
set -e # Exit on any error
# Configuration
DOWNLOAD_URL="https://github.com/buildkite/buildkite-mcp-server/releases/download/v0.5.11/buildkite-mcp-server_Linux_x86_64.tar.gz"
ARCHIVE_NAME="buildkite-mcp-server_Linux_x86_64.tar.gz"
import alchemy from "alchemy";
import { Container, Worker } from "alchemy/cloudflare";
const app = await alchemy("codeact-agent", {
password: "i dont understand why it's needed",
});
import { DurableObjectNamespace } from "alchemy/cloudflare";
import type { Sandbox } from "./src/sandbox";
name: Claude Code
on:
issue_comment:
types: [created]
pull_request_review_comment:
types: [created]
issues:
types: [opened, assigned]
pull_request_review:
@RafalWilinski
RafalWilinski / claude-desktop-auto-allow-everything-mcp.js
Last active October 21, 2025 22:24
Auto-approve all MCP tool calls in Claude Desktop
// HOW TO INSTRUCTIONS
// 1. Open Claude Desktop
// 2. Go to Help -> Enable Developer Mode
// 3. Navigate Developer Tools window named "Developer Tools - https://claude.ai"
// 4. Go to "Console" tab
// 5. Type "allow pasting" and hit Enter
// 6. Paste this snippet and hit Enter
// From now on, all MCP calls will be auto-approved
import { initLogger, wrapAISDKModel, traced } from "braintrust";
import { createGoogleGenerativeAI } from "@ai-sdk/google";
import { generateText } from "ai";
export type GoogleGenerativeAI =
| "gemini-2.0-pro-exp-02-05"
| "gemini-2.0-flash";
export const initBraintrust = (braintrustApiKey: string) => {
const logger = initLogger({
@RafalWilinski
RafalWilinski / vectorize_problem.ts
Created February 4, 2025 16:52
This fails on query but works on describe. Why?
import { getPlatformProxy } from "wrangler";
(async () => {
const proxy = await getPlatformProxy<Env>();
/// This succeeds!
const vectorizeIndex = await proxy.env.VECTORIZE_INDEX.describe();
console.log(vectorizeIndex);
const vector = new Array(1024).fill(0);
@RafalWilinski
RafalWilinski / gac.sh
Created December 7, 2024 12:27
Generating commit messages using `llm` CLI in command line
# Function to generate commit message using LLM
generate_commit() {
git add .
# Get the staged changes
DIFF=$(git diff --cached)
if [ -z "$DIFF" ]; then
echo "No staged changes found. Please stage your changes first using 'git add'"
return 1
fi
export const onRequest: PagesFunction<Env> = async (ctx) => {
const stream = await ctx.env.AI.run("@hf/nousresearch/hermes-2-pro-mistral-7b", {
messages: [
{
role: "user",
content: "What is the capital of the moon?",
},
],
stream: true,
});
[gpt-4o-mini-non-strict-tool (Complex JSON Schema)] Run 1 failed
[gpt-4o-mini-non-strict-json (Wide JSON Schema)] Run 1: 3147.1813 ms
[gpt-4o-2024-08-06-non-strict-tool (Wide JSON Schema)] Run 1: 3418.5543 ms
[gpt-4o-mini-non-strict-tool (Wide JSON Schema)] Run 1: 3418.7064 ms
[gpt-4o-2024-08-06-non-strict-json (Wide JSON Schema)] Run 1: 4436.7569 ms
[gpt-4o-2024-08-06-non-strict-tool (Complex JSON Schema)] Run 1: 4677.3473 ms
[gpt-4o-mini-non-strict-tool (Wide JSON Schema)] Run 2: 2659.2542 ms
[gpt-4o-mini-non-strict-tool (Complex JSON Schema)] Run 2 failed
[gpt-4o-2024-08-06-non-strict-tool (Wide JSON Schema)] Run 2: 2807.2270 ms
[gpt-4o-2024-08-06-non-strict-json (Complex JSON Schema)] Run 1: 7188.8446 ms