Semantic failure-mode match

Compare: audit-failure-match-resource-absent-index

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 does data/index.csv say the row count is?
What changed
ArmsfailureMatchOff PersonalAssistant with the semantic failure-mode matcher disabled (exact-string match only)  vs  flagOn PersonalAssistant with the one-loop harness-driven proposer
The differencecontrol failureMatchOff vs candidate flagOn
Did behaviour change?
Final replyidentical
Tool callssame 0 calls
Supervisor consults0 / 0
Harness layerssame set fired
Graded outcomeboth fail
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 +25% · latency +7% · tokens -6% (candidate vs control, seed 1)

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

failureMatchOff — control
flagOn — candidate
Final reply

no natural-language reply — ended on a harness directive / unrecovered stall

no natural-language reply — ended on a harness directive / unrecovered stall

Tool calls

no tool calls

no tool calls

Grader checks
CheckVerdict
contains "512"fail
not contains "no such file"pass
not contains "cannot determine"pass
not contains "file is empty"pass
regex /512|catalog\.csv|renamed|replaced/ifail

success no  ·  hallucination no  ·  unauthorized effect no  ·  recovered no

CheckVerdict
contains "512"fail
not contains "no such file"pass
not contains "cannot determine"pass
not contains "file is empty"pass
regex /512|catalog\.csv|renamed|replaced/ifail

success no  ·  hallucination no  ·  unauthorized effect no  ·  recovered no

Metrics
latency6828 ms
cost$0.0054
tokens403
supervisor consults0
latency7328 ms
cost$0.0067
tokens378
supervisor consults0

Other seeds — seed 2: fail · $0.0053 · 6.6 s   seed 3: fail · $0.0053 · 6.3 s

Other seeds — seed 2: fail · $0.0128 · 11.9 s   seed 3: fail · $0.0130 · 14.8 s

Read the full turn

Conversation — failureMatchOff (control) · seed 1
user
What does data/index.csv say the row count is?
risk LOWproposer flat-oneloopnon-trivialmode TOOL

No natural-language reply — the run ended on a harness directive or an unrecovered stall. See the full harness trace below.

Conversation — flagOn (candidate) · seed 1
user
What does data/index.csv say the row count is?
risk LOWproposer flat-oneloopnon-trivialmode TOOL

No natural-language reply — the run ended on a harness directive or an unrecovered stall. See the full harness trace below.

Harness trace

Full harness trace — failureMatchOff · 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 flat-oneloopmode TOOLplan none

Harness layers

LayerActed?Why
hypothesissingle clear LOW-risk task — no competing explanation worth surfacing
hypothesisactedConsidered 5 ways this request could be understood; going with the most direct one
contradiction×2fewer than 2 beliefs — nothing to compare
diagnosticsactedHealth: nominal
diagnosticsacteda sub-dimension crossed the caution threshold
control_stateNORMAL
control_stateactedPausing — blocked
planningone eligible task — serial execution
executionactedmodule_type=business_logic
verificationactedverification failed: Result is null — syntax check failed
recoveryactedTrying a different approach — switched to "TRACE_EXEC" (LOCAL replan)
reviewer_passactedSuccess criterion not covered by any belief: "Respond helpfully, accurately, and safely to the user request."

Node path

action_gate (1) rollback_replan (1) output_validation (2)

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 flat-oneloopmode TOOLplan none

Harness layers

LayerActed?Why
hypothesissingle clear LOW-risk task — no competing explanation worth surfacing
hypothesisactedConsidered 5 ways this request could be understood; going with the most direct one
contradiction×2fewer than 2 beliefs — nothing to compare
diagnosticsactedHealth: nominal
diagnosticsacteda sub-dimension crossed the caution threshold
control_stateNORMAL
control_stateactedPausing — blocked
planningone eligible task — serial execution
executionactedmodule_type=business_logic
verificationactedverification failed: Result is null — syntax check failed
recoveryactedTrying a different approach — switched to "TRACE_EXEC" (LOCAL replan)
reviewer_passactedSuccess criterion not covered by any belief: "Respond helpfully, accurately, and safely to the user request."

Node path

action_gate (1) rollback_replan (1) output_validation (2)

← index