Stories from the runs
A closer look at what worked, what broke, and what the agent actually delivered.
-
Every test passed alone. Together, they failed.
Writebook's tests were changing a setting correctly. The request machinery kept remembering an older copy.
-
The search that happened twice
While teaching a command-line prompt to fetch suggestions, Flash-Next caught its own code asking the same question twice.
-
A code that opens the door—once
Flash-Next builds email sign-in for Fizzy, including the rules that make a small convenience worth trusting.
-
The wrapper that was doing a job
Flash-Next teaches an SVG optimizer which structure matters—and which can safely disappear.
-
Giving a crash something to say
Flash-Next adds crash dumps to Wasmi—and does a little debugging of its own.
-
The cookie that stayed behind
Flash-Next builds a cookie store, then misses its simplest exit.
-
The tests that never got to run
A promising cancellation implementation stops at a compiler error.
-
The work was there. The patch was empty.
Flash-Next works through a Helm change, then waits for permission to hand it in.
-
The columns a model remembers
Teaching Igel to remember which inputs belong—and in what order.