inwo. Headroom sits between your AI agent and the model, compressing tool outputs, logs, RAG chunks, files and history before they cost you tokens — reversibly, and locally. What it is, how the router-plus-compressors design works, what its self-reported benchmarks actually show, and the honest costs.
agency-agents is an open-source collection of 218 specialized subagent 'personalities' you can install into Claude Code and a dozen other tools. What it is, how to install and use it, how it compares to writing your own, and the honest caveats.
Career-Ops is an open-source, agentic job-search pipeline that scores offers, tailors your CV, scans portals, and tracks everything — locally, with you in the loop. What it is, how to run it, how it compares to the alternatives, and the honest caveats.
Ponytail is a one-file 'lazy senior dev' skill that makes coding agents stop and pick the simplest solution that works — cutting code by ~54% (and tokens ~22%) on a real agentic benchmark. What it is, how the ladder works, what trusted reviewers found, and the honest caveats.
Ralph is a brutally simple technique — loop the same prompt into a coding agent until the task is done. What it is, how the two main implementations (Anthropic's ralph-wiggum plugin and snarktank/ralph) actually differ, the anecdotal numbers behind the hype, and an honest critical review of where it breaks.
What agent skills are, how Claude Code loads them (and what they cost in context), the real risks, and a lean, honest stack for code review, security, quality and multi-agent work — with install/remove/update for each.
OpenSpec is an open-source, lightweight spec-driven-development tool that adds a spec layer so humans and AI agree on what to build before code. What it is, how the propose → apply → archive workflow and delta specs work, how it compares to Spec Kit and Kiro, and its honest tradeoffs.
pi is a tiny, aggressively-extensible terminal coding harness. What it is, how to install and use it, how it compares to opencode and Claude Code, and what the benchmarks actually say.
Spec Kitty is an open-source Python CLI that turns product intent into a repo-native, spec-driven workflow — spec, plan, tasks, review, accept, merge — with isolated git worktrees for parallel agents, a local kanban dashboard, and governance gates. What it is, how it works, use cases, and where it fits versus lighter tools.
Two open-source takes on spec-driven development for AI coding agents — Spec Kitty's governed software factory versus OpenSpec's lightweight spec layer. The real differences, pros and cons, examples, and when to reach for each.
A practical look at LangChain — what it's for, how to install and use it, what you can build, whether you need to deploy it, how it stacks up against Google ADK and other agent frameworks, and its real downsides.
A community-built, Apache-2.0 library of 754 structured cybersecurity skills that give an AI agent senior-analyst playbooks — mapped to MITRE ATT&CK, NIST CSF 2.0, ATLAS, D3FEND and NIST AI RMF. What it is, how the skills are built, defensive use cases, how it compares, and the honest tradeoffs (including its dual-use responsibility).
Addy Osmani's Agent Skills packs the workflows, quality gates, and Google engineering practices that senior engineers use into markdown skills an AI coding agent follows across the whole development lifecycle. What it is, how it works, and the honest tradeoffs.
mattpocock/skills is Matt Pocock's everyday .claude skills folder, published — small, composable agent skills that fix the common failure modes of coding agents (misalignment, verbosity, broken code, ball-of-mud architecture) without taking over your process. What it is, how it works, the standout grilling and TDD skills, and the honest tradeoffs.
A single CLAUDE.md file, derived from Andrej Karpathy's observations on LLM coding pitfalls, that distills four principles — think before coding, simplicity first, surgical changes, goal-driven execution — to make Claude Code (and Cursor) behave. What it is, how it works, and an honest read on its viral metrics.
UI UX Pro Max is an open-source AI skill that gives a coding agent a design brain — a reasoning engine that turns a one-line prompt into a complete design system with styles, palettes, typography, and anti-patterns. What it is, how it works, what it produces, and the honest tradeoffs.
A plain-English explainer of the AI harness — the code around a model that lets it use tools, take steps, and get work done. What it is, why it matters, how it works, and the two senses of the word (agent harness vs evaluation harness).
Andrej Karpathy's LLM Wiki pattern — let an LLM build and maintain a persistent, interlinked wiki from your sources, so knowledge compounds instead of being re-derived on every query. What it is, how it works, how it compares to RAG, and the token economics.
A Claude Code skill that makes the agent talk like a caveman — cutting output tokens by up to ~75% while keeping full technical accuracy. Why, how it works, how to install it, and real before/after numbers.
Superpowers is an open-source plugin that gives a coding agent a full development methodology — brainstorm, plan, then build with TDD and subagent review. What it is, how it works (and why it stays token-light), real opinions, use cases, and its pros and cons.