ECC

a self-contained markdown prompt with YAML frontmatter declaring when it should auto-fire. Third-party writeups describe it as a reference implementation of Anthropic's SKILL.md format. affaan-m/everything-claude-code (~175k stars; plus a second repo affaan-m/ECC at ~205k stars which is the same project under a renamed identifier). 48 agent definitions, 182 SKILL.md files, 68 legacy slash-command shims, hooks, rules, MCP configurations, npm packages (ecc-universal, ecc-agentshield), a Tkinter desktop dashboard, and a security scanner (1282 tests, 102 static analysis rules). Includes per-harness adapters for Claude Code, Codex CLI, Codex macOS app, Cursor, OpenCode, Gemini CLI, and Antigravity. Anthropic hackathon winner. Three orders of magnitude in scope under one word. What's underneath is Anthropic's SKILL.md format: markdown with YAML frontmatter, auto-loaded at session start. The frontmatter declares when the skill should fire; the harness picks relevant skills based on the description and injects only those into context. It's RAG-over-prompts using model-based routing on descriptions rather than vector stores. The format works well enough that you can mix skills from different authors in the same .claude/ folder without the harness caring, which is the actual reason this took off. Trending packs ship per-harness adapters on top of that substrate so the same skill content installs into Codex, Cursor, OpenCode, etc., with per-harness rewrites. The trending list is measuring three different things under one label: small high-leverage CLAUDE.md edits (karpathy-skills, cost-to-try wins), curated personal reference sets (mattpocock, distribution-by-reputation wins), and full framework distributions (ECC, comprehensive-catalog-marketing wins). Stars are not telling us which of these are surviving on actual reuse. The defensibility implications are uncomfortable. When a startup pitches "our agent does X better because of our prompting and workflow," and the artifact is a folder of markdown files with YAML frontmatter, that's a Notion template, not a moat. Karpathy's four rules will be absorbed into the default behavior of the next Claude release. mattpocock's TDD skill is sharp but copyable. ECC's 182-skill catalog is impressive engineering, but the prompts inside can be diffed and ported in an afternoon. What does seem to hold value after reading these: harness ergonomics (install paths, hook plumbing, cross-tool sync scripts, MCP-server lifecycle), distribution (mattpocock-the-person is a moat, the markdown isn't), and security tooling around skill files specifically (AgentShield, ECC's scanner, is a real product even if the skills it scans aren't). None of those are the prompt. Repos: github.com/forrestchang/andrej-karpathy-skills, github.com/mattpocock/skills, github.com/affaan-m/everything-claude-code.

ce" to have the time necessary to learn and use AI a lot. Lots of different techniques that quickly became completely obsolete today. Without LLMs, it would have been extremely hard to have the same app than I have today. I used Flutter (Dart) to avoid having to dev and maintain two codebases. It is not a language I knew. Learning the language first would have severely hindered the process. For me, from copy/paste to using MCP then Roo Code, then Claude Code was an ecstatic process. I always loved having ideas but the time it would take me to build the thing and test it always felt too long. Not anymore. So we carefully designed, iterated and implemented the two codebases for Jynx. One for the flutter app, one for the firebase backend. I chose Firebase to avoid having to maintain a server and be able to focus on the UI/UX of the app. We started thinking about it in December 2024 and started devs early 2025; not working on it full time at all. We really poured our heart into it and we truly tried to make it as secure as possible. Even though we mean business, it is a passion project. By using the excuse of learning agentic flows, I took the time to inspect each aspects of the app's systems thoroughly. Tech stack: - Flutter 3.41 / Dart 3.11 (single codebase, iOS + Android) - Firebase (Firestore, Cloud Functions in TypeScript, Auth, Storage, FCM) - Riverpod 3.1 + Freezed + json_serializable for state management & immutable models - Drift for encrypted local SQLite caching (offline-first architecture to optimize Firebase costs) - Clean Architecture with feature modules and mixin-based repositories - Sentry + Firebase Crashlytics for production error reporting - Freerasp for runtime app self-protection (tamper detection, root/jailbreak) Agentic engineering artifacts: - Claude Code (Claude + GLM) as primary coding agent - 22 hooks, 18 skills, 13 instincts, 8 rule files, custom subagents, slash commands, MCP servers and plugins (instincts system from Affaan's https://github.com/affaan-m/everything-claude-code) - GitNexus - MemPalace for persistent context across sessions Stats: 1,239 Dart files, 214k lines of code (excluding generated boilerplate), 30k lines of comments across the Flutter codebase. I made a detailed cheatsheet document about my whole setup if you want it. I could post it or you DM me. If you have questions, ask away, I'll gladly answer. Test it and tell me what you think of it honestly, I won't get offended! Take care, Antoine