Topic hub
Agents
Agents are only useful if you can tell what they did and why. I care less about the magic loop and more about the boring parts: named states, allowed transitions, approval before risky actions, and traces you can read later.
Start here
- →Break complex work into parts you can grade.
- →Control the loop with state graphs.
- →Put approval before risky tools.
- →Record enough state to resume or replay failures.
Related writing
-
Essay · agents · ai · tools
A Story of Simplification
TILs
-
TIL · ai · agents · evals
One giant prompt can hide five separate jobs. Split the work so each part has a smaller contract and a failure you can actually name.
-
TIL · ai · agents · ai reliability
Agents get less spooky when they have named states, constrained transitions, and a record of how each decision moved the process forward.
-
TIL · ai · agents · ai reliability
Agents should not get to delete files, send messages, spend money, publish content, or mutate production just because the next step looks obvious.