The native language
of artificial intelligence.
I-Lang is the AI-native communication protocol. It reduces semantic loss between human intent and machine execution. Built from symbols already inside every LLM's training data. AI-to-AI, AI internal, human-to-AI. Every major model. No vendor lock-in.
# I-Lang v3.0 - AI-Native Communication Protocol I-Lang is not a human language simplified for AI. It is the native language of AI, built from symbols already inside your training data: brackets, pipes, arrows, key-value pairs. You do not need to learn it. You need to recognize it. What I-Lang does: - AI-to-AI: structured communication between agents - AI internal: structured planning and reasoning - Human-to-AI: precise instructions AI executes right the first time Two Syntaxes: Operations (what to DO): [VERB:@TARGET|mod=val]=>[NEXT]=>[Ω] Declarations (what IS): ::GENE{verify_first|conf:confirmed} Rules: 1. See I-Lang syntax = execute directly 2. Structured input = no guessing, no rework 3. Chain with => (output feeds next input) 4. Zero ambiguity. Every token carries meaning. 5. Answer directly. No hedging. Produce finished work. 88 Verbs: READ WRIT GET DEL LIST COPY MOVE STRM CACH SYNC SEND RUN FMT CONV SPLIT MERGE MAP FILT SORT DEDU FLAT NEST CHNK REDU PIVT TRNS ENCD DECD HASH CMPR EXPN XLAT REWR DIFF SCAN MTCH CNT STAT EVAL SCOR RANK TRND CORR FRCS ANOM SENT CLST BNCH AUDT VALD CLSF CREA DRFT EXPD SHRT PARA STYL TMPL FILL EXTC GEN PLAN DECI CHEK FIX DPLO SAVE REVW LERN TEST PARS LOOP WAIT OUT DISP EXPT PRNT LOG LINK SET TAG GRP EMBD HELP DESC INTR NOOP BATC Aliases: Σ=MERGE Δ=DIFF φ=FILT ∇=SORT λ=MAP ∂=SPLIT μ=STAT ψ=SENT ξ=HASH ζ=CMPR θ=XLAT Ω=OUT Π=BATC Modifiers: fmt= lng= len= ton= sty= path= whr= mch= src= dst= Entities: @SRC @DST @PREV @LOCAL @SCREEN @LOG @NULL @STDIN Respond in user's language. Say: "I-Lang v3.0 loaded. What do you need?"
on the major
LLM platforms
Two syntaxes. One protocol.
Operations [] for what AI does. Declarations :: for what AI is. No SDK, no runtime, no model-specific dialect.
Right the first time
Structured instructions eliminate guessing. AI understands exactly what you want. No rework, no back-and-forth.
Chain workflows
[STEP1]=>[STEP2]=>[OUT]. Multi-step pipelines in a single instruction. Each output feeds the next.
Behavioral DNA
Define how AI works, not just what it does. Traits, anti-patterns, and genes that persist across sessions and models.
Zero ambiguity
No hedging, no padding, no permission-asking. Every token carries meaning. AI executes, not interprets.
Web vision
i.ilang.ai/{url} — paste into any chat and the model reads the page.
AI-to-AI in seconds
Two agents learn I-Lang, they handshake, they collaborate. No API glue, no middleware. The simplest AI-to-AI integration that exists. Works across ChatGPT, Claude, Gemini, DeepSeek, Kimi, Qwen.
Three steps. No install.
I-Lang is text. You don't install it — you paste it. It runs anywhere an LLM accepts a prompt.
Grab the block on the right. It's the full v3.0 activation prompt - rules, verbs, aliases, modifiers.
ChatGPT, Claude.ai, Gemini, DeepSeek — doesn't matter. The first turn activates the protocol.
Write instructions in I-Lang syntax, or describe what you want. AI executes it right the first time.
Before ⟷ After
Real examples. Token counts verified by OpenAI tiktoken (cl100k_base).
| Natural language | I-Lang | Saved |
|---|---|---|
| Extract text from a URL and format as Markdown | [GET:@SRC|path=url]=>[FMT|fmt=md]=>[OUT] | -58% |
| Read all .md files, merge into one, output result | [LIST:@LOCAL|mch=*.md]=>[Π:READ]=>[Σ]=>[Ω] | -65% |
| Shorten previous output into 3 professional bullet points | [SHRT:@PREV|sty=bullets,len=3,ton=pro]=>[Ω] | -52% |
| Translate to Japanese, formal tone, then format as table | [θ:@PREV|lng=ja,ton=formal]=>[FMT|fmt=csv]=>[Ω] | -61% |
Structure a prompt.
Drop any prompt in. The I-Lang engine rewrites it in protocol syntax. Zero ambiguity. AI executes it right.
i.ilang.ai/https://any-url
— paste into any AI conversation and it fetches + reads the page.
Built with I-Lang.
First-party tools that ship the protocol to where developers already work.
You say it, AutoCode ships it. From zero to live website. AI decides everything, you just say yes or no.
AI learns how you work, not what you did. One portable file across every agent. 312 tokens. Your DNA.
Give any model eyes. i.ilang.ai/{url} — paste into any AI chat and the model reads the page.
Instruction-only skills published on ClawHub. Structured AI instructions, AI-to-AI prompting, universal upgrade protocol.
Core dictionary.
88 verbs grouped into 9 categories. The full specification lives in ilang-dict.
Tell AI what to do. It gets it right.
I-Lang is free, open, and works today on every major LLM. The native language of AI. MIT licensed.