#430 · Primary category: AI Coding Assistants
TabNine
AI Code Completions
Project last updated:09/04/25
GitHub Stars
10.8K
Forks
538
Contributors
19
License
MIT
Why we included this project
Most of what makes TabNine tick lives behind a closed door: the backend is closed source, so this repository contains no source files at all. What it does hold is the protocol the engine uses to talk to editor plugins, plus the configuration files that define how languages are tokenized and which file extensions belong to the same language. A client sends JSON requests over stdin and receives completions on stdout, and files like languages.yml and language_tokenization.json control the details, so identifiers from .c files get suggested in .h files, while a dash can be part of an identifier in Lisp but not in Java. If you want to write your own TabNine client, or just study how a language-aware completion engine handles these boundaries, this is a useful reference. Teams evaluating the service should know the engine itself is proprietary; this repo is a spec-and-config companion, not runnable software you can fork.
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.