#241 · Primary category: AI Coding Assistants
magic-context
Unbounded context. Memory that manages itself. One session, for life. The hippocampus for coding agents, part of CortexKit.
Project last updated:08/29/26
GitHub Stars
1.9K
Forks
100
Contributors
27
License
MIT
Why we included this project
Anyone who has run a long session with a coding agent like OpenCode or Pi knows the wall: the conversation grows until the model starts forgetting what it said an hour ago, and the next task begins with a blank slate. Magic Context sits underneath that agent and keeps a compact, tiered history of older turns, so the session keeps going without the usual lossy compaction pauses. It also remembers durable project facts, rules, and constraints across sessions, which means the agent still knows your conventions weeks later. The context management is prompt-cache aware, so it works with provider caching instead of re-billing your prefix. If you are tired of re-explaining your codebase every morning, this is a practical way to turn your assistant into a long-term teammate.
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.