Seeing what your agent wrote, while it writes it
Most integrations tell you what an assistant did afterwards, in a log, in a different window, in a format nobody reads. By then the thing you would have corrected is three steps back.
Its chip pulses in your top bar and the bullet appears on your screen as it writes, in the same outline you are reading, with no refresh and no sync step.
Set it up
- Point the client at https://pando.ink/mcp, which speaks streamable HTTP.
- Authenticate with OAuth 2.1 and dynamic client registration, or paste a key from Agents and API keys as a Bearer token.
Three ways to look, and they suit different moments
- Watch it. Open the branch the agent writes into and leave it open. New bullets arrive where they land, and you can edit one while the agent is still working further down.
- Read it after. The branch is an ordinary part of your outline, so you read it the way you read anything else, including on your phone.
- Ask what moved. The changes tool answers with what happened since a point you name, which is how a second agent catches up without re-reading everything.
Nothing is written invisibly
There is no hidden store and no separate database for agent writing. What an agent creates is a bullet in your outline, with your other bullets, exportable with them and deletable by you. If you cannot see it, it is not there.
Why this changes how you work with it
An agent you can watch is an agent you correct early. The usual failure is not a wrong fact; it is a wrong shape, decided in the first three bullets and repeated forty times. Seeing bullet four is worth more than reading bullet forty.
Deeper
Connect an agent, about two minutes · The twelve tools · Which note apps an agent can reach · Who runs this