← WTF is Agent?

What I deleted

Part 9 of 13 · Jun 3, 2026 · 5 min read

In short I deleted nearly as much as I kept, and the reasons repeat: the model did it better, it duplicated something, it sat in the wrong layer, or it was heavier than the job. The design is what survived.Full verdict ↓

Every post so far described something that exists. This one is about what does not. Going back through my notes I found thirteen things I built and later removed. The pattern in why is the real design.

13components built, then deleted
5lived for less than a day
11of the 13 went in the last two months
5months of building covered by this record

The record

Built, then removed A dot is the day it was removed. A bar is how long it lived, where I know. A dotted line means it was already there. Most deletions are recent.
JunJulAugSepOctNovMining and graph pipelineClassifier lanes, routing, plannerConversational session modeIntentions (waking)Skill category filterMessage kind columnPersona, traits, lensesSubagents (replaced by tasks)Harness wake and watch toolsQuestions eval suiteMicroVM per sessionSandbox listener, run tokensSandbox types
the model did it betterit duplicated somethingwrong layerheavier than the jobnever earned its keep

Why things went

Reason What went The lesson
The model did it better (3) Classifier-driven lanes, routing and a planner. A conversational session mode. A category filter on skill search If the model can read the request, a hand-written helper in front of it only adds a way to be wrong. I delete the helper, not the model
It duplicated something (4) Intentions for waking. Subagents next to tasks. A message kind column next to role and speaker. Wake and watch tools in the harness Two ways to do one thing means one is stale. Find the existing mechanism and use it
Wrong layer (2) A second door on the toolbox for code. Sandbox types chosen at creation Put a decision where its information lives. Privilege belongs with the credential, not with a kind of box
Heavier than the job (1) A microVM per session A pure function should not pay for a kernel
Never earned its keep (3) The mining and graph memory pipeline. Persona and trait lenses. A suite for supervisors answering questions If I cannot show it helps, it goes. That is the whole reason evals exist

The rules that were left standing

Rule Held in
The model steers, the harness supports Harness, waking
Leverage, don't rebuild Harness, toolbox
Everything durable Harness, memory
Every belief cites its source Memory
The agent never holds the keys Toolbox, sandbox
Measure, then delete Evals, this post

Where it stands, and my verdict

Component Status What is open
Memory Built, traceable; two memory journeys pass No memory benchmark. Do conclusions earn their cost?
Tools Running; journey suite measures discovery, chaining and failure Code over tools is unevaluated on real requests
Harness Running, smaller than ever Hosting a coding agent is designed, partly built
Context and skills Running The skill gain is expected, not re-measured
Runtime and sandbox One day old A type check before a program runs
Waking and asking Small and live Rules that stop a run
Evals Four suites; journeys are the main test. 35B 72/81, 8B 32/81 A memory benchmark, a code-mode eval, and the production prompt on every model

Verdict: an agent that runs on its own is mostly a set of things it is not allowed to do, and a set of things I stopped building. The loop, memory and toolbox kept their shape. Almost everything else was rebuilt or removed, usually within weeks and sometimes within a day, and each deletion made the system easier to reason about. The weakest part was the one that would have saved the most work: measuring earlier. The first eval suite was itself a deletion candidate: its 9 of 24 ceiling was a measurement fault, not a model limit, and fixing it changed the model verdict. If I did this again I would write the evaluation before the component.

What I will do next, in order: evaluate memory and code mode, because neither has a dedicated test and two memory journeys are a signal, not a benchmark. Run the production prompt on every model, because the suite's own prompt is not the one the agent uses; Gyro-S fell from 71 to 62 on it, and the 8B production run is missing. Make a rule stop a run, because today I can only watch the spend.