Launching today
Contextify - ctxfy.com
Scale Your Context. Keep Your State.
5 followers
Scale Your Context. Keep Your State.
5 followers
Ctxfy is a Context State Engine that compresses LLM conversation history by up to 95%. It strips chit-chat and redundancy, preserving only "Hard State" — decisions, constraints — into a portable State Object and Artifacts - code, schemas, etc. Drop it into any model as a system prompt, inject it into agent loops, or fast-forward it with new diffs and logs. REST API, model-agnostic, zero-retention mode available.
I'm Oleg — solo founder and the only engineer behind Ctxfy.
How it started: I had hundreds of LLM conversations I didn't want to lose. Not just transcripts — real thinking. Decisions I'd worked through, architectures I'd debated, ideas I'd refined over hours of back-and-forth. Buried in chat windows, impossible to search, impossible to reuse. All that value, just sitting there rotting.
I didn't want to summarize them manually. I wanted to extract the meaning — the hard logic, the decisions, the constraints — and carry it forward into the next conversation. Or the next model. Or the next project.
So I built a tool to do exactly that. Then someone wanted an API. Then agent integrations. Then infinite context loops for autonomous pipelines. What started as "I don't want to lose my chats" turned into a full Context State Engine compatible with Openai GPT, Claude, Gemini, and Llama.
Stunning, isn't it? ✨
What Ctxfy does:
- Up to 95% token reduction — strips chit-chat, keeps decisions, code, constraints
- /fast-forward — inject new diffs or docs into a frozen state
- Infinite agent memory — use Ctxfy as a garbage collector for long-running loops
- Zero-retention mode for sensitive workloads
Being a solo dev means I personally read every piece of feedback. If something is broken, confusing, or missing — I want to know. If you're building something and think Ctxfy could fit your stack, I'm genuinely open to integrating, collaborating, or just jumping on a call to think through your use case.
Free tier is live today — no credit card needed.
What's your biggest headache with LLM context right now? I'm all ears. 👇