#agent-design
Every post tagged "agent-design".
112 posts · showing 21–30
-
Your Agent Has One manage_calendar Tool. That's Why It Deleted the Wrong Meeting.
A single do-everything tool looks tidy in the schema and hides four separate problems: guessed actions, unvalidated payloads, permissions that can't tell reading from deleting, and traces that flatten three risk levels into one name. Here's where to cut instead — and where splitting makes things worse.
-
Map-Reduce or Sequential Refine? How an Agent Should Read 400 Pages
Chunking 400 pages of contracts forces a choice most teams make by accident. One shape quietly drops evidence as it goes; the other can never spot a conflict between two documents — and the fix is not the pattern, it's what the map step returns.
-
Who Decides the Agent Is Done? Three Stop Conditions, and Why "The Model Said So" Isn't One
Most first agents end their loop when the model stops calling tools — which means the model, not you, decides what "done" means. Here are the three stop conditions a real loop needs, and the twenty lines that turn a silent wrong answer into another iteration.
-
What Crosses the Handoff: Transcript, Summary, or Contract?
A triage agent spends eleven tool calls finding the bug, then hands off. Passing the transcript ships its dead ends; passing a summary strips the uncertainty. What a typed handoff contract gets right — and the one field teams always forget.
-
The Tool Returned 200 and Nothing Happened
A calendar agent reported success on a meeting that landed at 3am, on the wrong calendar, with no invitation sent. The API did exactly what it was asked — and the return value gave the agent no way to know.
-
Let the Model Pick the Transition, Not the Tool
A returns agent that can issue a refund before it has checked eligibility will eventually do exactly that, and no amount of prompt wording reliably stops it. Making the call unavailable is a design decision, not a prompt one.
-
Extraction Was Never the Hard Part: An Invoice Agent, Rebuilt
The first version had the model read invoices and type the numbers. The rebuild started after the numbers were already read — which is where the actual work in accounts payable has always been.
-
Four Tool Calls to One Answer: A Research Run, Annotated
A research agent answers one question in four steps. Reading what changed in the transcript between each step explains more about how agents work than any diagram of the loop.
-
The Supervisor That Should Have Been a For-Loop
Every night the same plan, a model call per dispatch, and about once a week it checked eleven services instead of twelve. A for-loop cannot skip an item; an agent orchestrating one eventually will.
-
Designing an Agent You Can Redirect Mid-Run
Twelve files into forty, you realize the approach is wrong. Let it finish or kill it — both bad, and the third option had to be designed in before the run started.