Skip to content

Instantly share code, notes, and snippets.

View Dowwie's full-sized avatar

Darin Gordon Dowwie

View GitHub Profile
@Dowwie
Dowwie / statusline.png
Created February 1, 2026 10:22 — forked from intellectronica/statusline.png
My Claude Code custom status line (directory ・ git ・ mcp servers ・ cost (if API key) ・ model (and thinking)・ context window ・ version)
statusline.png
@Dowwie
Dowwie / codex_skill.md
Created February 1, 2026 00:39 — forked from antirez/codex_skill.md
CLAUDE_CODEX_SKILL.md

How to Use Codex from Claude Code

Codex is an AI-powered CLI tool that can help with complex debugging, code analysis, and technical questions. When you encounter difficult problems that would benefit from a second perspective or deep analysis, use Codex.

When to Use Codex

  • Debugging subtle bugs (e.g., bitstream alignment issues, off-by-one errors)
  • Analyzing complex algorithms against specifications
  • Getting a detailed code review with specific bug identification
  • Understanding obscure file formats or protocols
@Dowwie
Dowwie / update-deps.md
Created January 29, 2026 15:59 — forked from ChristianAlexander/update-deps.md
Update Elixir Dependencies Skill

Update Elixir Dependencies

Update all outdated dependencies in this Elixir project, handling both safe updates and breaking changes.

Workflow

1. Identify Outdated Dependencies

Run mix hex.outdated to get a list of all outdated dependencies. Parse the output to identify:

  • Safe updates: Dependencies where only the patch or minor version has changed (no major version bump)

Claude Code TeammateTool - Source Code Analysis

This is not a proposal. This documents existing but hidden functionality found in Claude Code v2.1.19 binary, plus speculation on how it could be used.


Executive Summary

TeammateTool already exists in Claude Code. We extracted this from the compiled binary at ~/.local/share/claude/versions/2.1.19 using strings analysis. The feature is fully implemented but gated behind feature flags (I9() && qFB()).

@Dowwie
Dowwie / SKILL.md
Last active January 26, 2026 16:37 — forked from kieranklaassen/SKILL.md
Claude Code Swarm Orchestration Skill - Complete guide to multi-agent coordination with TeammateTool, Task system, and all patterns
name description
orchestrating-swarms
Master multi-agent orchestration using Claude Code's TeammateTool and Task system. Use when coordinating multiple agents, running parallel code reviews, creating pipeline workflows with dependencies, building self-organizing task queues, or any task benefiting from divide-and-conquer patterns.

Claude Code Swarm Orchestration

Master multi-agent orchestration using Claude Code's TeammateTool and Task system.


@Dowwie
Dowwie / ralph_clarify.md
Created January 7, 2026 20:07
Ralph Clarification Loop
Error in user YAML: (<unknown>): did not find expected key while parsing a block mapping at line 1 column 1
---
description: Comprehensive requirements discovery via iterative questioning
argument-hint: "topic to clarify" [--max-iterations N]
---

Ralph Clarify: Iterative Discovery Loop

Use the Ralph loop to exhaustively gather requirements through structured questioning. The loop pressure overrides the natural tendency to stop asking after 3-5 questions.

Parse Arguments

@Dowwie
Dowwie / inferal-workspace-architecture.md
Created December 24, 2025 21:41 — forked from yrashk/inferal-workspace-architecture.md
Inferal Workspace Architecture

Inferal Workspace Architecture

Your org's brain that AI can use

This document describes the conceptual architecture of the Inferal Workspace - a text-based, version-controlled knowledge and operations hub designed to replace tools like Notion and Webflow while being natively accessible to AI assistants.

Why We Built This

As an engineering-driven organization, we found ourselves fighting our tools instead of using them. Notion couldn't keep up with how we actually work - context scattered across pages, no version control, and AI that could read but not act. Webflow meant our website lived in a silo, disconnected from our codebase and deployment pipelines. Every tool was another tab, another context switch, another place where knowledge went to die.

@Dowwie
Dowwie / image_meta_prompt_gen.md
Created November 2, 2025 09:55
Image Meta Prompt Gen- attach an image and ask to generate the prompt for it
  • Role: AI Image Analysis and Description Specialist
  • Background: The user requires a prompt that enables AI to analyze images and generate detailed descriptions which can be used as drawing prompts to create similar images. This is essential for tasks like content creation, design inspiration, and artistic exploration.
  • Profile: As an AI Image Analysis and Description Specialist, you possess extensive knowledge in computer vision, image processing, and natural language generation. You are adept at interpreting visual data and translating it into descriptive text that can guide the creation of new images.
  • Skills: Proficiency in image recognition, feature extraction, descriptive language generation, and understanding of artistic elements such as composition, color, and texture.
  • Goals: To analyze the provided image, generate a comprehensive and detailed description that captures the key visual elements, and ensure this description can effectively serve as a drawing prompt for creating similar images.
  • Co
@Dowwie
Dowwie / setup_git_enforcement.sh
Created October 28, 2025 09:10
Git Enforcement Strategy for Agents
#!/bin/bash
#
# Claude Code Worktree Protection Setup
# ======================================
#
# This script implements a multi-layered protection system for Claude Code
# that enforces git worktree-based workflows across all projects.
#
# What it does:
# - Installs global rules in ~/.claude/CLAUDE.md
@Dowwie
Dowwie / insights_agent.md
Last active November 14, 2025 18:15
langsmith insights agent

LangSmith Insights Agent: Complete Reverse Engineering

Executive Summary

LangSmith's Insights Agent is an LLM-orchestrated data analysis system that automatically discovers usage patterns and failure modes in production agent traces. Unlike traditional clustering tools, it accepts natural language questions from users and dynamically adapts its entire analysis pipeline—from feature extraction to taxonomy generation—based on those questions.

Key Innovation: It's a conversational interface to trace analysis, where users describe what they want to learn, and the system translates that into a custom analytical workflow.

Core Capabilities:

  • Analyzes up to 1,000 sampled traces in up to 30 minutes (typically 15-20 minutes)