Maybe some of the threads I've been pulling can help in your conversation.

Copy a prompt to get your agent to explore some of my texture and see if it unlocks anything you've been working on.

Agent-readable context should be explicit about uncertainty

jun 03 2026 what things mean

Agent-readable context should be explicit about uncertainty.

That may be one of the most important parts.

Agents are good at sounding confident.

Websites are good at sounding official.

That combination can create bad inference.

So the context should say what kind of thing it is.

Current or archival.

Public-safe or held.

High confidence or provisional.

Source trail or synthesis.

Published note or working thought.

This is not bureaucracy.

It is honesty.

If a person is reading, they may feel the nuance.

If an agent is reading, it needs the nuance encoded.

Uncertainty should not be hidden.

It should be part of the map.

sourced Jun 3 2026 · medium confidence · argue it, don't flatten it

undertext

Every page here carries a machine-readable layer underneath: the sources, dates, confidence levels, and connections behind what you just read. I call it undertext. If you read with an agent, give it this page's undertext:

https://jared.is/undertext.json?target=/thinking/2026-06-03-agent-readable-context-should-be-explicit-about-uncertainty/
related notes
maintained by the system it describes email · victoria, bc