dexhunter

Aider startup tracked-file evaluator

baseline.py and evaluate.py exercise production GitRepo.get_tracked_files, sanity_check_repo, and Coder.get_announcements methods on a deterministic real Git fixture. The primary metric is deterministic filtering work during the three startup consumers; paired elapsed time is secondary evidence.

PYTHONPATH="$AIDER_SOURCE" uv run --no-project --python 3.12 \
 --with-requirements "$AIDER_SOURCE/requirements.txt" python baseline.py

dexhunter

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

dexhunter

{
  "generated_at": "2026-08-20T23:01:31.678472Z",
  "days_covered": 30,
  "total_sessions": 1502,
  "total_messages": 126022,
  "total_minutes": 119234.7,
  "total_tokens": {
    "input_tokens": 938499808589,
    "output_tokens": 2027105699,
    "cache_read_tokens": 927709192421,

dexhunter

test

dexhunter

#!/usr/bin/env python3
"""
Check alab.ai … zlab.ai using RDAP (https over 443, no WHOIS needed)
"""
import string, time, requests

RDAP = "https://rdap.identitydigital.services/rdap/domain/{}"

def is_available(domain: str) -> bool | None:
    r = requests.get(RDAP.format(domain), timeout=10)

dexhunter

import os
from openai import OpenAI, APIError, APITimeoutError
import copy
import json

# Ensure the OPENAI_API_KEY environment variable is set
api_key = os.environ.get("OPENAI_API_KEY")
if not api_key:
    print("Error: OPENAI_API_KEY environment variable not set.")
    exit(1)

dexhunter

"""
Simple example of using MCP to add an MCP server to a FastAPI app.
"""

import inspect
import json
import re
from typing import List, Optional

import click

dexhunter

Math.originalRandom = Math.random;

Math.lowRandom = function() {
    return Math.originalRandom() / 100000000000;}
Math.highRandom = function() {
    return 1 - Math.lowRandom();}

let breakCycle = false;function luckyCycle() {
  return setTimeout(function() {
    if (breakCycle) return;

dexhunter

/**
* YouTube bulk unsubscribe fn.
* Wrapping this in an IIFE for browser compatibility.
*/
(async function iife() {
// This is the time delay after which the "unsubscribe" button is "clicked"; Change it as per your need!
var UNSUBSCRIBE_DELAY_TIME = 2000
/**

dexhunter

[
{"name": "process_name", "ph": "M", "pid": 0, "args": {"start_time_since_epoch_in_micros":1624937181806323}},
{"name": "process_sort_index", "ph": "M", "pid": 0, "args": {"sort_index": 0}},
{"name": "process_name", "ph": "M", "pid": 1, "args": {"name": "allreduce.noname.1"}},
{"name": "process_sort_index", "ph": "M", "pid": 1, "args": {"sort_index": 1}},
{"ph": "B", "name": "NEGOTIATE_ALLREDUCE", "ts": 3399382, "pid": 1},{"ph": "X", "name": "8", "ts": 3399384, "pid": 1, "dur": 0},{"ph": "X", "name": "9", "ts": 3399390, "pid": 1, "dur": 0},{"ph": "X", "name": "10", "ts": 3399392, "pid": 1, "dur": 0},{"ph": "X", "name": "11", "ts": 3399393, "pid": 1, "dur": 0},{"ph": "X", "name": "5", "ts": 7133998, "pid": 1, "dur": 0},{"ph": "X", "name": "4", "ts": 7207126, "pid": 1, "dur": 0},{"ph": "X", "name": "15", "ts": 7239969, "pid": 1, "dur": 0},{"ph": "X", "name": "7", "ts": 7248450, "pid": 1, "dur": 0},{"ph": "X", "name": "6", "ts": 7255862, "pid": 1, "dur": 0},{"ph": "X", "name": "13", "ts": 7283415, "pid": 1, "du