Foundations

Foundations



Primary Sources




© TinyFish, Inc.. All rights reserved.




Primary Sources




Every agent deployment estimate runs the same arithmetic. Per-step success rate, raised to the number of steps, watch the floor open. Directionally right, and nobody should be argued out of it. But it borrows two things the evidence hasn't supplied: one per-step rate to raise, and steps that fail independently of each other. Five runs of the same task, same agent, same instruction, different bill. Horizon is doing most of the damage in the reliability results, and it is the variable those studies measure least directly.
Every agent deployment estimate runs the same arithmetic. Per-step success rate, raised to the number of steps, watch the floor open. Directionally right, and nobody should be argued out of it. But it borrows two things the evidence hasn't supplied: one per-step rate to raise, and steps that fail independently of each other. Five runs of the same task, same agent, same instruction, different bill. Horizon is doing most of the damage in the reliability results, and it is the variable those studies measure least directly.
A basket, a second item, a card that only the person at the keyboard can produce, and a refund that may take a day to land. Four calls, one story.
The 2026-07-28 revision removed the connection that used to hold that story together. Nothing was actually subtracted. The handshake, the session, the durable job each have a successor, and every one of those successors now sits on your side of the line. What follows is what a shopping cart looks like once the protocol stops keeping it for you.
A basket, a second item, a card that only the person at the keyboard can produce, and a refund that may take a day to land. Four calls, one story.
The 2026-07-28 revision removed the connection that used to hold that story together. Nothing was actually subtracted. The handshake, the session, the durable job each have a successor, and every one of those successors now sits on your side of the line. What follows is what a shopping cart looks like once the protocol stops keeping it for you.
A wrong parameter on a single tool call is annoying. A lost constraint propagating through a five-step chain is dangerous. A compressed summary that quietly drops the one fact preventing a catastrophic action twenty steps later is something else entirely. The failure modes change character as tasks get longer, not just frequency.
Recent research quantifies what practitioners already feel in their debugging sessions: capability gains have outpaced reliability gains by roughly seven to one, and failures compound nonlinearly with step count. Scaffolding that fixes one band's problems can introduce new failure surfaces at the next. These six cards map each step-count band to its dominant failure mode and the specific instrumentation check worth running this week.
A wrong parameter on a single tool call is annoying. A lost constraint propagating through a five-step chain is dangerous. A compressed summary that quietly drops the one fact preventing a catastrophic action twenty steps later is something else entirely. The failure modes change character as tasks get longer, not just frequency.
Recent research quantifies what practitioners already feel in their debugging sessions: capability gains have outpaced reliability gains by roughly seven to one, and failures compound nonlinearly with step count. Scaffolding that fixes one band's problems can introduce new failure surfaces at the next. These six cards map each step-count band to its dominant failure mode and the specific instrumentation check worth running this week.
Past Articles

An agent updates a billing tier, submits a regulatory form, merges a PR. Three days later someone shows up asking who au...

An agent runs a query. $14.2M comes back, clean execution, no errors. Looks like an answer. But the best-evaluated syste...

Agent outputs keep getting more detailed, more structured, more confident-looking. That's exactly the problem. A text-to...

Hallucination has a name. Refusal has a name. There's a third AI failure mode nobody's bothered to name yet: output that...