# Fylle — Content Intelligence for AI-Native Teams > Fylle (founded 2024, Milan) is an AI-native content intelligence > system. It structures company knowledge into a persistent, governable > memory layer for AI — with neural context graphs, compound intelligence > through feedback loops, and universal interoperability via MCP > (Model Context Protocol). The AI changes, the structured context stays. ## Product - [How Fylle Works](https://fylle.ai/#engine): Neural Structure, Compound Intelligence, Universal Interoperability — three capabilities no other platform combines - [.fylle Protocol](https://fylle.ai/protocol/): Open standard (Apache 2.0) for packaging, sharing, and deploying AI agents across any platform - [Agent Builder](https://fylle.ai/protocol/builder): Visual tool for creating .fylle agent packages — define identity, model, prompt, tools, guardrails and export - [Lehve](https://lehve.fylle.ai): LinkedIn intelligence tool for personal branding, powered by Fylle structured context ## Journal - [From Interaction to Action](https://fylle.ai/journal/from-interaction-to-action/): How people and machines turn a vague need into a precise act, and what changes with AI. - [Fylle Commons: Why We Opened Our Agents, and What Still Doesn't Travel](https://fylle.ai/journal/fylle-commons-why-we-opened-our-agents/): Fylle Commons shows that agent behavior is shareable, but durable advantage lives in private context - [What Can't Be Said Must Be Trained](https://fylle.ai/journal/what-cant-be-said-must-be-trained/): The line between context engineering and fine-tuning is verbalizability - [The 7% That Does the Work](https://fylle.ai/journal/the-7-percent-that-does-the-work/): A compact workspace inside language models carries disproportionate causal leverage - [AI's Value Already Shifted. We're Still Looking in the Wrong Place.](https://fylle.ai/journal/ai-value-already-shifted/): AI value is shifting from models to orchestration, but orchestration cannot fix thin context - [Il valore dell'AI si è spostato](https://fylle.ai/it/journal/il-valore-ai-si-e-spostato/): Original Italian version of AI's Value Already Shifted - [Context Engineering for Marketing](https://fylle.ai/journal/context-engineering-for-marketing/): Why context engineering beats prompt engineering as the durable marketing AI moat - [Context Engineering per il Marketing](https://fylle.ai/it/journal/context-engineering-per-il-marketing/): Italian translation of Context Engineering for Marketing - [The Moat Is Memory](https://fylle.ai/journal/the-moat-is-memory/): Software is moving from tools that forget to systems that remember - [Il vantaggio è la memoria](https://fylle.ai/it/journal/il-vantaggio-e-la-memoria/): Italian translation of The Moat Is Memory - [Agent Orchestration](https://fylle.ai/journal/agent-orchestration-marketing/): Why coordinating multiple AI agents is becoming a core marketing leadership skill - [Agent Orchestration: la Skill fondamentale per i Marketing Leaders](https://fylle.ai/it/journal/agent-orchestration-marketing/): Italian translation of Agent Orchestration - [Why Most AI Marketing Projects Fail](https://fylle.ai/journal/why-ai-marketing-projects-fail/): Why isolated AI tools fail without shared context and integration - [Perché la Maggior Parte dei Progetti AI di Marketing Fallisce](https://fylle.ai/it/journal/perche-progetti-ai-marketing-falliscono/): Italian translation of Why Most AI Marketing Projects Fail - [Vibe Marketing Is a Context Problem](https://fylle.ai/journal/vibe-marketing-context-problem/): Vibe marketing succeeds when the AI actually knows the brand - [Vibe Marketing è un Problema di Contesto](https://fylle.ai/it/journal/vibe-marketing-problema-di-contesto/): Italian translation of Vibe Marketing Is a Context Problem - [Infrastructure Declares Its Politics](https://fylle.ai/journal/infrastructure-declares-its-politics/): Why AI infrastructure choices about context ownership, portability and interoperability are political decisions - [L'infrastruttura dichiara la sua politica](https://fylle.ai/it/journal/infrastruttura-dichiara-la-sua-politica/): Italian translation of Infrastructure Declares Its Politics - [A Portable Format for Agents](https://fylle.ai/journal/a-portable-format-for-agents/): Why we built the .fylle protocol and why we gave it away as open source - [Un formato portabile per gli agent](https://fylle.ai/it/journal/un-formato-portabile-per-gli-agent/): Italian translation of A Portable Format for Agents - [Anthropic Just Leaked Our Expiration Date](https://fylle.ai/journal/anthropic-just-leaked-our-expiration-date/): How Anthropic's memory features validated our architecture thesis - [Anthropic ha svelato la nostra data di scadenza](https://fylle.ai/it/journal/anthropic-ha-svelato-la-nostra-scadenza/): Italian translation of Anthropic Just Leaked Our Expiration Date - [I Hate Reinventing the Wheel](https://fylle.ai/journal/i-hate-reinventing-the-wheel/): The Fylle foundation story — from content agency to AI infrastructure company - [Odio Reinventare la Ruota](https://fylle.ai/it/journal/odio-reinventare-la-ruota/): Italian translation of the Fylle foundation story ## Integrations - [MCP Server](https://mcpserver-production-6cd0.up.railway.app/mcp): Fylle's Model Context Protocol server — access structured context, cards, and briefs from Claude, ChatGPT, Cursor, or any MCP-compatible tool - [GitHub](https://github.com/FylleAI/.fylle): Open-source .fylle protocol repository ## Company - [About Fylle](https://fylle.ai/): Milan-based AI company, Fylle S.R.L., Via Bovisasca 85, 20157 Milano - [Contact](mailto:contact@fylle.ai): contact@fylle.ai - [LinkedIn](https://www.linkedin.com/company/fylle-ai): Company page # From Interaction to Action > How does something vague become precise? Description: From a vague need to a precise act: how people and machines move through the same levels, and why AI is the first machine that starts from the vague part. Published: 2026-09-28 Reading time: 6 min read Tags: ["signal"] --- Every action starts somewhere vague. A dry mouth. A page you need on paper. A few seconds later there's a glass in your hand, or a sheet coming out of a printer. In between, something vague becomes precise. This piece follows that path, level by level, first in a person, then in a machine. Then it looks at what changes now that AI can start from the vague part too. ## Where I started: clarity My starting point was Wittgenstein. The [Tractatus](https://gutenberg.org/ebooks/5740) opens with a promise: what can be said at all can be said clearly. [Russell](https://plato.stanford.edu/entries/logical-atomism/) and the logicians of his time wanted exactly that. A language with no ambiguity. In a way, that language exists now. We use it every day. It's called code. **But intentions start before words, and before words everything is vague.** Code only begins after someone has done the work of being precise. ## How it works for a person Take the simplest thing. You're thirsty, and you drink a glass of water. It looks like one gesture. It goes through five levels. | Person: five-node loop | What happens | |---|---| | Interaction | Dry mouth | | Intention | I want water | | Plan | That glass, there | | Action | Reach, adjust, drink | | Feedback | Thirst fades. The goal ends. | These levels aren't separate floors. They talk to each other all the time. [The body corrects while it moves](https://doi.org/10.1038/nn963), and sometimes the intention only takes shape while you're doing it. **The goal is yours.** Shaped by your body, your culture, other people. But yours. ## How it works for a machine Now the machine version. You press Print, and a few seconds later a sheet of paper comes out of a printer. | Machine: printing | What happens | |---|---| | Interaction | You press Print. The dialog specifies Pages: 1–3, Copies: 2, Printer: Office. | | Interpretation | Code writes instructions. | | Handoff | Print queue → Network → Sent to the printer. | | Execution | Ink on paper. | | Feedback | “Printed”, or “Paper jam”. | The print dialog does something important before anything else: it makes you be precise. Which pages, how many copies, which printer. The vague part stays with you. From there, each level translates for the next one. Code turns the document into instructions, the system sends them along, and at the end the printer moves. **The machine starts from a goal it received**, and settles every ambiguity before it acts. ## The ladder and the circle Same levels. Two architectures. | Comparison | Person | Machine | |---|---|---| | Where the goal comes from | From a need | Received | | When it decides | While acting | Before acting | | Where feedback reaches | Every level | Some levels | ### Where AI comes in With classic software, the hardest step is yours: turning a vague want into precise input. That's what the print dialog is for. A [language model](https://www.youtube.com/watch?v=LCEmiRjPEtQ) takes over that step. You say what you want in your own words, and the model does the interpretation that used to need a dialog, or a programmer. | Input | Interpretation | What follows | |---|---|---| | Classic software: what you want | You make it precise. | Same ladder below. | | With a language model: what you want, in your own words. “Print the pricing part for tomorrow, two copies.” | The model makes it precise. | Same ladder below. | **It's the first machine that starts closer to where people start.** The ladder below it stays the same, and the goal is still received. Agents are starting to [close the loop](https://www.figure.ai/news/helix) in a few places: they try, check, and correct. What's still missing is real learning during use, and goals of their own. ## The levels, one by one Scroll. The person and the machine go down together, level by level. The person and the machine descend through five levels. Feedback reaches every level for the person; machine feedback returns to Interpretation and Handoff, never to the goal. | Level | Person | Machine | |---|---|---| | 1 | **Interaction.** Your mouth is dry. Nobody asked you anything: your body did. | **Interaction.** You press Print. The dialog has already made you precise: which pages, how many copies. | | 2 | **Intention.** I want some water. The need becomes a goal. | **Interpretation.** The app turns the document into instructions the printer understands. | | 3 | **Plan.** That glass, on the table. The goal meets what's around you. | **Handoff.** The instructions travel through the print queue and the network to the printer. | | 4 | **Action.** You reach, grip, drink. Your hand adjusts all the way. | **Execution.** The printer moves the paper and puts ink exactly where it was told. | | 5 | **Feedback.** Thirst fades, and the intention simply ends. Or it doesn't, and you pour another. | **Feedback.** "Printed", or "Paper jam". It can try again. It can't decide you needed a different page. | ## An open question So, in this view, AI is a new disambiguation tool: it takes on our ambiguity and translates it into the deterministic level of code. For now, AI systems don't generate their own goals. They only work through short chains of tasks, and the ultimate goal still originates from an interaction with us. We've also seen that they haven't yet closed the loop. But if, as is happening, we're moving toward these systems being able to close the feedback loop at every level, making them adaptive and therefore more capable of generating genuine goals of their own, are we still talking about tools? --- ## Further reading - Ludwig Wittgenstein, *Tractatus Logico-Philosophicus* (1921): [gutenberg.org/ebooks/5740](https://gutenberg.org/ebooks/5740). Where this piece started: what can be said, and how clearly. - Kevin Klement, "Russell's Logical Atomism", Stanford Encyclopedia of Philosophy: [plato.stanford.edu/entries/logical-atomism](https://plato.stanford.edu/entries/logical-atomism/). Russell's idea of a language built from simple, unambiguous parts. - Edsger W. Dijkstra, "On the foolishness of 'natural language programming'" (EWD667): [cs.utexas.edu](https://www.cs.utexas.edu/~EWD/transcriptions/EWD06xx/EWD667.html). Why machines have always needed us to be precise. - Elisabeth Pacherie, "Action", Open Encyclopedia of Cognitive Science (MIT Press, 2025): [doi.org/10.21428/e2759450.3036d218](https://doi.org/10.21428/e2759450.3036d218). An accessible overview of how intentions turn into actions. - Emanuel Todorov and Michael I. Jordan, "Optimal feedback control as a theory of motor coordination", Nature Neuroscience (2002): [doi.org/10.1038/nn963](https://doi.org/10.1038/nn963). Why the body corrects while it moves. - Andrej Karpathy, "Software Is Changing (Again)", YC AI Startup School (2025): [youtube.com](https://www.youtube.com/watch?v=LCEmiRjPEtQ). Programming in plain language: the model as the new interpreter. - David Silver and Richard Sutton, "Welcome to the Era of Experience" (2025), overview by TechTalks: [bdtechtalks.com](https://bdtechtalks.com/2025/04/21/are-we-at-the-cusp-of-a-new-era-for-artificial-intelligence/). Agents that learn from their own stream of experience. - Figure AI, "Helix: A Vision-Language-Action Model for Generalist Humanoid Control" (2025): [figure.ai/news/helix](https://www.figure.ai/news/helix). A slow system that understands and a fast one that acts.