Voice has become the feature every AI agent platform rushes to ship. The obvious move is to build it as a standalone channel, something that sits beside your web app, your CLI tool, or your Telegram bot. It feels intuitive. You see voice, you create a voice interface. But that instinct creates a brittle architecture. It duplicates work, corrupts your logs, and slowly pulls your project context out of shape.
At APC and APX, we chose a different route. Voice is not a channel. It is a mode. It sits on top of a surface rather than replacing it. Getting this distinction right is what keeps the system from drifting apart.
The Wrong Abstraction
When you treat voice as its own channel, you implicitly assume that speaking to an agent is a fundamentally different conversation from typing to one. Engineering teams respond by splitting the codebase. Suddenly there is a CLI channel and a separate voice-CLI channel. There is a web channel and a parallel voice-web channel. Each demands its own prompt variations, formatting rules, and context handling logic.
This is where the mess begins. A tweak to an agent's behavior must now be copied across multiple prompt trees. If the team forgets one surface, the experience fragments. Users get one tone over text and a slightly different personality over speech. Over time, these small inconsistencies compound into system drift. The portable context layer stops being portable because it has to account for vocal delivery in one branch and silent text in another. The abstraction leaks, and your once-unified project definition frays into a collection of channel-specific hacks.
Splitting Context from Runtime
To prevent this, we split responsibilities between two layers that stay strictly apart.
APC holds the project context. It defines the agents, the rules, and the skills that make up a project. Think of it as the stable meaning of the system. It answers the structural questions. What does this agent know? What is it allowed to do? What tools can it call? APC should remain completely agnostic about whether a reply is rendered on a screen, sent through a chat API, or piped through a speaker.
APX handles the runtime layer. It manages the surfaces you actually touch: the CLI, the web application, the desktop interface, the Telegram bot. When a user sends a request, APX chooses where and how to present the response. Deciding whether to format an answer for reading or optimize it for speaking is a runtime concern. It belongs in APX, not in APC.
This separation means a project defined in APC stays intact no matter how many surfaces APX exposes. The contract does not change. Only the presentation layer does.
How Modes Actually Work
In our implementation, surfaces like Telegram, the CLI, and the web app are channels. A channel tells you where an interaction happened. Voice is layered through channel metadata as a mode. A mode tells you how a reply should behave.
The prompt builder respects this boundary. It pulls from the project context in APC, then inspects the channel metadata. If the desktop surface is running in voice mode, the builder appends targeted instructions only at that moment. Maybe it cues the model toward shorter sentences, clearer punctuation for synthesis, or spoken number conventions. If the same desktop surface is running in text mode, those vocal instructions never enter the prompt.
The result is a single prompt tree per surface. There is no separate voice-desktop branch. There is no whisper-web variant. The modifier applies only when the runtime asks for it, and only at the last responsible moment. The core prompt stays constant.
What You Gain
This architecture pays off in three concrete ways.
Lower maintenance costs. If voice were its own channel, every surface would need a twin. You would maintain a CLI channel and a voice-CLI channel, a Telegram channel and a voice-Telegram channel, and so on. Every time you adjust a system prompt, fix a formatting bug, or refine a skill description, you would have to propagate that change across both trees. Miss one, and users notice the gap. By using a mode, you keep one prompt tree per surface. Voice becomes a conditional overlay rather than a fork in the road, so your workload stays linear as you add new ways to interact.
Logging accurato. I canali registrano dove è avvenuta un'interazione. Le modalità registrano come è stata erogata la risposta. Un'interazione desktop rimane un'interazione desktop, sia che l'utente la legga o la ascolti. Quando il tuo team traccia un bug o analizza i dati, non deve riconciliare "desktop-voice" con "desktop-text" come se fossero diverse superfici di prodotto. L'identificatore del canale rimane pulito e il flag della modalità si posiziona ordinatamente accanto ad esso nei metadati. I tuoi log rimangono fedeli e il debugging rimane semplice perché posizione e comportamento non sono intrecciati tra loro.
Contesto di progetto pulito. APC definisce il contratto. Non dovrebbe importargli se una risposta viene pronunciata, sussurrata o renderizzata con un font monospace. Queste sono questioni di runtime. Mantenendo la formattazione vocale all'interno di APX, preserviamo la portabilità di APC. Puoi prelevare una definizione di progetto APC e inserirla in un ambiente di runtime completamente nuovo senza trascinare con te assunzioni di formattazione specifiche per la voce o scorie di ottimizzazione del parlato. Il confine regge e il significato del progetto rimane stabile.
La prova sul Desktop
Il nostro percorso desktop lo dimostra nell'uso quotidiano. Il desktop è la superficie. Quando un utente abilita la voce, il sistema esegue quella stessa superficie desktop in modalità vocale. Poiché la voce risiede nello strato delle modalità, il canale desktop mantiene il suo intero contesto e comportamento. Non diventa un prodotto diverso con regole diverse. Il prompt builder si limita a notare il flag e aggiunge istruzioni vocali solo quando necessario. Quando l'utente torna al testo, quelle istruzioni scompaiono completamente. Il contesto del progetto sottostante non è mai cambiato. Il desktop è sempre stato il desktop.
Il punto fondamentale
L'idea centrale è semplice. APC descrive il significato stabile del progetto. APX descrive l'esecuzione a runtime. La voce è un modificatore di una superficie, non una sua sostituzione. Trattala in questo modo e i tuoi prompt rimarranno piccoli. I tuoi log rimangono chiari. I tuoi