What AI Agents does with a Markdown file, and where an image fits

AI Agents takes Markdown through a file sitting in the working directory the agent is allowed to read, and that is the door this page is about. The question worth asking is what an image looks like by the time it arrives, because AI Agents never sees the image itself. It sees whatever the conversion handed over, and nothing else.

AI Agents reads a heading tree as a map of where to look next, and skips whatever the map does not mention. What comes out of an image here is the text of the image as Markdown, with a table rebuilt when the image held one, so the structure AI Agents is looking for is actually present instead of merely implied. That is the entire trick, and there is nothing clever underneath it.

What survives from an image into AI Agents

The conversion drops the background, the stamps, the shadows and the skew of the scan before AI Agents ever sees them, because none of it carries meaning into AI Agents and all of it costs something. What it keeps is the part you would otherwise have retyped by hand.

• The text comes out in reading order rather than in scan order, and AI Agents receives it as structure rather than as something to infer.
• A visible table is rebuilt as a Markdown table instead of collapsing into a line of words, which is the part AI Agents would otherwise have to reconstruct on its own.
• References and figures are read character by character, which is the part people retype by hand, so nothing downstream of AI Agents has to guess at it.

What AI Agents never gets back from an image

Every conversion costs something, and a page listing only the gains is a sales page. Here is what an image gives up on the way to AI Agents, stated plainly enough that you can decide against it.

• Layout and colour are gone, and a colour that meant something meant it only to a human, and AI Agents will not get it back from anywhere.
• Anything too small or too blurred to read is not returned, and not flagged either, so nothing you build in AI Agents should depend on it.
• Certainty is gone: a model is reading, and a model can be wrong while sounding sure, which is a real cost inside AI Agents and not a rounding error.

The Image to Markdown step behind this AI Agents page was measured

Gemini 2.5 Flash Lite does the conversion, and it was proven end to end on 30 August 2026: a purchase order image built for the occasion, carrying a unique reference, came back with the reference, the quantity and the unit price all read correctly. That is one run on one day and not an average, which is exactly why the date travels with the claim instead of being left out of it. The same output is what AI Agents would have received.

This capability returned nothing at all until 30 August 2026, and two probes failed that same day before the quota was restored. AI Agents has no way of telling you any of that from the inside, so it gets said here instead, on the page that sent you there.

Once the image is Markdown, what to do inside AI Agents

With the Markdown in hand, keep one Markdown file a subject rather than one long file, because an agent pays for everything it opens. That is a AI Agents habit rather than a conversion setting, and it is where most of the value of converting an image in the first place actually lands.

The constraint worth knowing before you start: an agent re-reads the same file at every step it takes, so the size of that file is a recurring cost. It bites the same way whether the Markdown came out of an image or out of something else entirely, but it bites hardest on long documents, which is what an image usually is.

What we will not pretend about Image in AI Agents

• AI Agents will act on a mangled sentence rather than stop and ask about it, and converting an image first does not change that in the slightest.
• a loop that reads a noisy file ten times pays for that noise ten times, which is a AI Agents behaviour rather than anything the image conversion did.
• The conversion costs half a credit and the price is shown before the image is sent, whether the Markdown ends up in AI Agents or nowhere at all.