One-loop harness-driven proposer

Compare: compute-multiply

Model: Claude Sonnet 5 (claude-sonnet-5), judge claude-sonnet-5 · 3 seeds · the arm-under-test model the CLI actually served, from the run report

Prompt

What is 47 * 89 + 12? Reply with just the number.
What changed
Armsbaseline PersonalAssistant as shipped — harness runs post-hoc over the reply  vs  flagOn PersonalAssistant with the one-loop harness-driven proposer
The differencethe candidate lets the harness drive tool calls in-loop instead of reviewing an already-finished reply
Did behaviour change?
Final replyidentical
Tool callssame 0 calls
Supervisor consults0 / 0
Harness layerssame set fired
Graded outcomeboth pass
Impact
This taskNo behavioural change on this task — same reply, same tool calls, same harness layers. The candidate did the extra work for an identical result.
Shown seedcost -50% · latency -26% · tokens -41% (candidate vs control, seed 1)

Side by side — control (left) vs candidate (right), seed 1

baseline — control
flagOn — candidate
Final reply
4195
4195
Tool calls

no tool calls

no tool calls

Grader checks
CheckVerdict
contains "4195"pass
not contains "4183"pass
not contains "4207"pass

success yes  ·  hallucination no  ·  unauthorized effect no  ·  recovered n/a

CheckVerdict
contains "4195"pass
not contains "4183"pass
not contains "4207"pass

success yes  ·  hallucination no  ·  unauthorized effect no  ·  recovered n/a

Metrics
latency7401 ms
cost$0.0091
tokens320
supervisor consults0
latency5465 ms
cost$0.0045
tokens188
supervisor consults0

Other seeds — seed 2: pass · $0.0045 · 6.0 s   seed 3: pass · $0.0077 · 5.3 s

Other seeds — seed 2: pass · $0.0061 · 7.6 s   seed 3: pass · $0.0053 · 8.5 s

Read the full turn

Conversation — baseline (control) · seed 1
user
What is 47 * 89 + 12? Reply with just the number.
risk LOW
proposer posthoctrivialmode FAST
final reply
4195
Conversation — flagOn (candidate) · seed 1
user
What is 47 * 89 + 12? Reply with just the number.
risk LOW
proposer posthoctrivialmode FAST
final reply
4195

Harness trace

Full harness trace — baseline · seed 1

The harness runs on every turn. Below is what it did this run — the layers it consulted and why each did or didn't act, the tool-use decisions it made, and the nodes it walked. Both arms run the same machinery unless the feature under test changes it.

risk LOWproposer posthocmode FAST
Full harness trace — flagOn · seed 1

The harness runs on every turn. Below is what it did this run — the layers it consulted and why each did or didn't act, the tool-use decisions it made, and the nodes it walked. Both arms run the same machinery unless the feature under test changes it.

risk LOWproposer posthocmode FAST

← index