#436 · Primary category: AI Coding Assistants
ctx
Repo-aware recommendations for skills, agents, MCP servers, and model harnesses. Use your own inventory or the shipped 79,958-node graph with 68,494 skills, 467 agents, 10,790 MCPs, and 207 harnesses.
Project last updated:08/24/26
GitHub Stars
581
Forks
71
Contributors
5
License
MIT
Why we included this project
Deciding which skills, MCP servers, or agent harnesses to load for a repository is usually guesswork. ctx's CTX Fit command turns that into a test: it profiles your repo, builds representative tasks from your history, and runs candidate AI coding configurations against them in an isolated, network-disabled sandbox. The cheapest setup that reliably passes your own test command wins, chosen by a fixed rule (reliability floor first, then cost, then simplicity), so an LLM can explain a result but never decide it. That is useful for teams adopting AI coding tooling who want evidence over vendor claims and do not want to spend on configurations that do not fit their codebase. The free local profile is read-only and spends nothing, so trying it costs nothing, and the paid evaluation path hands you a reviewable diff or pull request before anything is applied.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
superpowers
An agentic skills framework & software development methodology that works.
ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
skills
Skills for Real Engineers. Straight from my .agents directory.
opencode
The open source coding agent.
andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.