The conducting · Asking
Where you actually use it: a chat window, a command line, or the agents you already run. All three reach the same gate.
01
The plainest way in. Open the portal, ask, and the question goes straight to the gateway, the same one that governs every other route in. Nothing about the chat window makes the answer softer or the check looser. It's a window onto the same gate everything else uses.
02
If you'd rather type than click, there's a command line for that. It talks to the same gate, the same rules. Today it works against a copy of your Mind on the machine you're running it from. A version that reaches your Mind from anywhere is being built.
03
This is the one worth pausing on. The agents and tools you already run, the ones doing research, drafting, or scheduling on your behalf, can connect to the same gateway you do. They don't get a side channel with looser rules. They get the same gate, the same checks, and a receipt issued to that agent by name. When one of them asks your Mind something, you can see exactly what it asked and what it was allowed to read, same as if you'd asked it yourself.
04
You can walk in through chat, the command line, or an agent acting on your behalf. Three doors. One gate. One score behind it. The answer doesn't shift depending on which door you used, because there's only one Mind back there answering, not three copies drifting apart.
I got tired of re-explaining the same context to five different tools, each one holding its own half-true version of things. I wanted one answer, the same one, no matter which app I'd opened that morning.
ANTONY · FOUNDER
FILED
running since 2026-05 · chat and connected agents reach the gate today · a remote command line is being built, the local version works now
NEXT Skills and agents What runs on a Mind, and what holds it in check →The founding cohort is small on purpose. Each Mind is brought up personally.
MENSARA.AI · ASKING · MACHINE SURFACE · TEXT/MARKDOWN
# Asking > A Mind can be reached through three surfaces: the chat portal, a command line for technical owners, and the owner's own agents and tools connecting as authenticated clients. All three route through the same gateway and get the same governed, receipted treatment, so the answer doesn't change depending on which surface asked. Canonical: https://mensara.ai/how-it-works/asking ## Chat, in the portal, wherever you are The plainest way in. Open the portal, ask, and the question goes straight to the gateway, the same one that governs every other route in. Nothing about the chat window makes the answer softer or the check looser. It's a window onto the same gate everything else uses. ## A command line, for people who live there If you'd rather type than click, there's a command line for that. It talks to the same gate, the same rules. Today it works against a copy of your Mind on the machine you're running it from. A version that reaches your Mind from anywhere is being built. ## The agents and tools you already run This is the one worth pausing on. The agents and tools you already run, the ones doing research, drafting, or scheduling on your behalf, can connect to the same gateway you do. They don't get a side channel with looser rules. They get the same gate, the same checks, and a receipt issued to that agent by name. When one of them asks your Mind something, you can see exactly what it asked and what it was allowed to read, same as if you'd asked it yourself. ## One Mind behind all of them You can walk in through chat, the command line, or an agent acting on your behalf. Three doors. One gate. One score behind it. The answer doesn't shift depending on which door you used, because there's only one Mind back there answering, not three copies drifting apart. **Status:** running since 2026-05 · chat and connected agents reach the gate today · a remote command line is being built, the local version works now ## Prompt ```prompt Connect one of your own agents or automations to your Mind as an MCP client using your existing gateway credentials. Have it ask a question you could also ask in the chat portal, then compare the two receipts. Same principal type. Same verdict. Same zones touched. ``` ## Related - [The gateway](https://mensara.ai/how-it-works/the-gateway) - [Receipts](https://mensara.ai/how-it-works/receipts)