Skip to content

Instantly share code, notes, and snippets.

View jeremyronking's full-sized avatar

Jeremy King jeremyronking

View GitHub Profile
@jeremyronking
jeremyronking / spec-writer.SKILL.md
Created September 20, 2026 01:11
spec-writer — Claude Code skill: interactive spec writer for LLM implementing agents, with a mandatory ambiguity lint that gates issue creation
name spec-writer
description Interactive specification writer that captures requirements and creates focused, phased implementation plans with requirements as GitHub issues or local spec files
argument-hint
feature or bug description

Purpose

Generate focused, precise implementation specs through an interactive Q&A process, breaking features into logical phases and creating specs in dependency order.

Audience: the specs are consumed by LLM implementing agents (of varying models and capabilities), not humans. Different models fill spec gaps differently — wherever a spec is ambiguous, each implementer invents a different answer. Precision converges them; narrative prose does not. Every rule below serves that goal: exact contracts, machine-checkable done-ness, explicit anti-goals, and zero undefined qualifiers.

@jeremyronking
jeremyronking / compose-work.SKILL.md
Created September 20, 2026 01:11
compose-work — Claude Code skill: front-load every decision, then turn a feature description into GitHub issues and queued agent tasks hands-off
name compose-work
description End-to-end workflow that spec-writes a feature (using /spec-writer and /grill-me), creates GitHub issues, and queues Composer agent tasks. Accepts a description or existing GitHub issue URL.
argument-hint
description or GitHub issue URL

Purpose

End-to-end workflow that takes a feature idea from specification to a queued Composer agent task. Combines spec-writing (with /grill-me for clarity), GitHub issue creation, and Composer MCP task creation into a single orchestrated flow.

Key principle: Front-load ALL user interaction, then execute autonomously. Gather every decision (requirements, task type, model, effort) before creating issues or tasks. Once the user gives final approval, the entire creation pipeline runs hands-off.

@jeremyronking
jeremyronking / daily-update.sh
Created May 24, 2026 18:00
Launchd brew and npm updates
#!/bin/zsh
#
# daily-update.sh — keep Homebrew packages and global npm packages up to date.
#
# ─── What this script does ─────────────────────────────────────────────────
# 1. brew update — refresh Homebrew's package metadata
# 2. brew upgrade — upgrade installed formulae/casks
# 3. brew cleanup — remove old versions and caches
# 4. npm update -g — update global npm packages within their semver ranges
# (does NOT bump major versions; that requires
@jeremyronking
jeremyronking / statusline.sh
Last active August 6, 2026 16:52
Claude Code Status Line - Enhanced statusline showing model, mode, context usage, token counts, session cost, API limits with reset times, and git branch
#!/bin/bash
# ==============================================================================
# Claude Code Status Line
# ==============================================================================
# This script creates a rich status line for Claude Code showing:
# - Current model being used
# - Mode (if in plan/edit mode)
# - Context window usage percentage
# - Token usage (input/output totals for the session)
# - Session cost in USD