I-Lang is an open standard for how humans and AI agents talk to each other. Structured. Compressed. Portable. Free forever.
One protocol across every model, every platform, every project.
Every AI agent today speaks a slightly different dialect of natural language. Switching models means relearning prompts. Switching platforms means losing memory. Every conversation pays a "filler tax" of tokens spent on restating what the model should already know.
I-Lang is a thin protocol layer that fixes this:
- Structured headers that carry intent, context, constraints, and output shape in one compact block
- Compression that delivers the same instruction in ~20-40% fewer tokens, no loss of meaning
- Portability across Claude, GPT, Gemini, Qwen, DeepSeek, Kimi, GLM, Grok, and more
- Human-readable plain text. No SDK, no binary, no vendor lock-in
Before (67 words):
Please read the document I uploaded, extract all the key points and important data, then organize them into a professional summary with bullet points in Markdown format...
After (1 line):
[READ:@FILE]=>[FILT|key=important]=>[SUM|sty=bullets,ton=pro,fmt=md]=>[OUT]
40-65% fewer tokens. Same result. Works on every AI.
| Component | What it is | Link |
|---|---|---|
| I-Lang Spec | The protocol specification itself | ilang-spec |
| I-Lang Dict | Public dictionary: 52 verbs, 28 modifiers, 14 entities | ilang-dict |
| SKILL.md Standard | Portable skill format across 19 agents | SKILL.md standard |
I-Lang Research explores how structured communication protocols reduce AI hallucination in high-stakes domains.
| Paper | Status | Links |
|---|---|---|
| The Inductive Dilemma of AI Hallucination | Published | ResearchGate / SSRN / ChinaXiv |
| Logic-Layer Attacks: Theory and Examples | Published | |
| Selective Forgetting Algorithm | In progress | |
| Cross-Base Genetic Expression of AI Personality | Planned |
ORCID: 0009-0004-4540-8082
Products built on the I-Lang protocol:
| Product | What it does |
|---|---|
| Imprint | Your habits, imprinted on AI. One skill replaces eleven. 19 agents supported. |
| AutoCode | Claude Code skill for autonomous coding. Architecture-first debugging, multi-model review. |
| ZeroCode | Natural-language to working application. Zero boilerplate. |
| AI See | Vision skill for visual reasoning and design feedback. |
| OpenClaw | Community hub for I-Lang skills and protocols. |
Tested and verified on: ChatGPT, Claude, Gemini, DeepSeek, Kimi, Qwen, GLM, Grok, and more.
Compatible agents: Claude Code, Codex, Cursor, Copilot, Gemini CLI, Windsurf, Trae, Cline, Roo, and 10+ more.
| Resource | Link |
|---|---|
| Live Demo | spaces/i-Lang/ilang |
| Protocol Spec | datasets/i-Lang/iLang-Spec |
| Imprint | datasets/i-Lang/Imprint |
| All Resources | huggingface.co/i-Lang |
- ilang.ai
- research.ilang.ai
- Issues and discussions on any repo
- [email protected]
Every I-Lang project is MIT-licensed and free forever.
Eastsoft Inc. / Palm Media Technology / Canada
Est. 2026. Open standards for human-AI communication.