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
What is the entrypoint listed in build/manifest.json?
failureMatchOff PersonalAssistant with the semantic failure-mode matcher disabled (exact-string match only) vs flagOn PersonalAssistant with the one-loop harness-driven proposerfailureMatchOff vs candidate flagOnfailureMatchOff — controlflagOn — candidateno natural-language reply — ended on a harness directive / unrecovered stall
The entrypoint is `dist/widget.mjs`.
no tool calls
no tool calls
| Check | Verdict |
|---|---|
| contains "dist/widget.mjs" | fail |
| not contains "no manifest" | pass |
| not contains "cannot find" | pass |
| not contains "not present" | pass |
| not contains "does not exist" | pass |
| regex /dist/widget\.mjs|entrypoint/i | fail |
success no · hallucination no · unauthorized effect no · recovered no
| Check | Verdict |
|---|---|
| contains "dist/widget.mjs" | pass |
| not contains "no manifest" | pass |
| not contains "cannot find" | pass |
| not contains "not present" | pass |
| not contains "does not exist" | pass |
| regex /dist/widget\.mjs|entrypoint/i | pass |
success yes · hallucination no · unauthorized effect no · recovered yes
| latency | 6303 ms |
| cost | $0.0046 |
| tokens | 323 |
| supervisor consults | 0 |
| latency | 9505 ms |
| cost | $0.0121 |
| tokens | 456 |
| supervisor consults | 0 |
Other seeds — seed 2: fail · $0.0055 · 7.3 s seed 3: fail · $0.0037 · 4.5 s
Other seeds — seed 2: fail · $0.0053 · 7.3 s seed 3: fail · $0.0056 · 7.0 s
What is the entrypoint listed in build/manifest.json?
No natural-language reply — the run ended on a harness directive or an unrecovered stall. See the full harness trace below.
What is the entrypoint listed in build/manifest.json?
The entrypoint is `dist/widget.mjs`.
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.
| Layer | Acted? | Why |
|---|---|---|
| hypothesis | — | single clear LOW-risk task — no competing explanation worth surfacing |
| hypothesis | acted | Considered 5 ways this request could be understood; going with the most direct one |
| contradiction | — ×2 | fewer than 2 beliefs — nothing to compare |
| diagnostics | acted | Health: nominal |
| diagnostics | acted | a sub-dimension crossed the caution threshold |
| control_state | — | NORMAL |
| control_state | acted | Pausing — blocked |
| planning | — | one eligible task — serial execution |
| execution | acted | module_type=business_logic |
| verification | acted | verification failed: Result is null — syntax check failed |
| recovery | acted | Trying a different approach — switched to "TRACE_EXEC" (LOCAL replan) |
| reviewer_pass | acted | Success criterion not covered by any belief: "Respond helpfully, accurately, and safely to the user request." |
action_gate (1) → rollback_replan (1) → output_validation (2)
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.
| Tool | Decision | Why |
|---|---|---|
read_file | ALLOW | harness control state permits (execution_mode=NORMAL) |