TAP version 13 # Subtest: controlled engine HTTP freezes one child, pages original owned input/report, rejects foreign reads and prevents changed-source reuse ok 1 - controlled engine HTTP freezes one child, pages original owned input/report, rejects foreign reads and prevents changed-source reuse --- duration_ms: 17523.9393 type: 'test' ... # Subtest: native text and call IDs preserve original result pairing while unsupported content fails closed ok 2 - native text and call IDs preserve original result pairing while unsupported content fails closed --- duration_ms: 0.7573 type: 'test' ... # Subtest: one engine child freezes real installation identity before creating tasks and rejects a changed source ok 3 - one engine child freezes real installation identity before creating tasks and rejects a changed source --- duration_ms: 12099.2441 type: 'test' ... # Subtest: owned native tools return saved results, use real approvals/checks and account each actual request once across new engine Runs ok 4 - owned native tools return saved results, use real approvals/checks and account each actual request once across new engine Runs --- duration_ms: 28311.7459 type: 'test' ... # Subtest: forged native scope cannot change a parent file and a final completion claim cannot replace worker proof ok 5 - forged native scope cannot change a parent file and a final completion claim cannot replace worker proof --- duration_ms: 17543.3328 type: 'test' ... # Subtest: duplicate native IDs are rejected without replaying a saved tool ok 6 - duplicate native IDs are rejected without replaying a saved tool --- duration_ms: 15988.6919 type: 'test' ... # Subtest: native input exceeding the context budget is retained and sends no model request ok 7 - native input exceeding the context budget is retained and sends no model request --- duration_ms: 13494.8423 type: 'test' ... # Subtest: stop retains a late reported model usage once and executes no tool from that output ok 8 - stop retains a late reported model usage once and executes no tool from that output --- duration_ms: 14653.481 type: 'test' ... # Subtest: a source change after the original model intent stops before native tools execute ok 9 - a source change after the original model intent stops before native tools execute --- duration_ms: 14659.5568 type: 'test' ... # Subtest: an engine source change while approval is pending invalidates the old permission before any check starts ok 10 - an engine source change while approval is pending invalidates the old permission before any check starts --- duration_ms: 18136.5421 type: 'test' ... # Subtest: missing or noncompleted turn end cannot be adopted even with an owned receipt and assistant text ok 11 - missing or noncompleted turn end cannot be adopted even with an owned receipt and assistant text --- duration_ms: 0.4784 type: 'test' ... # Subtest: closed root with an original live descendant remains unknown and retains its process tree ok 12 - closed root with an original live descendant remains unknown and retains its process tree --- duration_ms: 0.2467 type: 'test' ... # Subtest: host recovery reserves original pending attempts exactly once and explicit review preserves unknown outcomes ok 13 - host recovery reserves original pending attempts exactly once and explicit review preserves unknown outcomes --- duration_ms: 16522.7044 type: 'test' ... # Subtest: corrupted model reservation cannot reduce recovered budget or permit a new engine request ok 14 - corrupted model reservation cannot reduce recovered budget or permit a new engine request --- duration_ms: 15420.2555 type: 'test' ... # Subtest: engine detection seals the actual CLI, SDK, profile and Node and rejects mixed versions without launch ok 15 - engine detection seals the actual CLI, SDK, profile and Node and rejects mixed versions without launch --- duration_ms: 686.4997 type: 'test' ... # Subtest: owned engine probe waits for real ready identity, verifies six native responses and saves immutable task evidence ok 16 - owned engine probe waits for real ready identity, verifies six native responses and saves immutable task evidence --- duration_ms: 1255.9979 type: 'test' ... # Subtest: an invalid startup nonce is rejected before initialize and preserves the process close and partial report ok 17 - an invalid startup nonce is rejected before initialize and preserves the process close and partial report --- duration_ms: 1232.4293 type: 'test' ... # Subtest: SDK identity and session directory mismatches retain raw responses and never become verified ok 18 - SDK identity and session directory mismatches retain raw responses and never become verified --- duration_ms: 3999.0792 type: 'test' ... # Subtest: cancel during a pending initialize closes only the owned process without sending an unsupported MCP cancellation ok 19 - cancel during a pending initialize closes only the owned process without sending an unsupported MCP cancellation --- duration_ms: 1462.171 type: 'test' ... # Subtest: cancel before readiness sends no protocol request and a user retry creates a different process and record ok 20 - cancel before readiness sends no protocol request and a user retry creates a different process and record --- duration_ms: 1862.3487 type: 'test' ... # Subtest: source or scratch changes and unexpected model traffic fail after preserving the actual observed result ok 21 - source or scratch changes and unexpected model traffic fail after preserving the actual observed result --- duration_ms: 4068.4306 type: 'test' ... # Subtest: unowned idle notifications stay quarantined and server requests receive a rejection without granting operations ok 22 - unowned idle notifications stay quarantined and server requests receive a rejection without granting operations --- duration_ms: 2316.2069 type: 'test' ... # Subtest: engine probe environment excludes parent secrets and NODE_OPTIONS while preserving operating system paths ok 23 - engine probe environment excludes parent secrets and NODE_OPTIONS while preserving operating system paths --- duration_ms: 1039.2714 type: 'test' ... # Subtest: foreign sessions and altered reports or task bindings cannot reuse verified engine evidence ok 24 - foreign sessions and altered reports or task bindings cannot reuse verified engine evidence --- duration_ms: 1288.897 type: 'test' ... # Subtest: restart never replays an active engine check or stops a historical PID, and keeps prior task state ok 25 - restart never replays an active engine check or stops a historical PID, and keeps prior task state --- duration_ms: 1361.0823 type: 'test' ... # Subtest: model argument validation and runtime activity do not grant external engine prompt or project execution ok 26 - model argument validation and runtime activity do not grant external engine prompt or project execution --- duration_ms: 532.2723 type: 'test' ... # Subtest: a follow-up question receives the actual failed engine check from its original Task and keeps its report unchanged ok 27 - a follow-up question receives the actual failed engine check from its original Task and keeps its report unchanged --- duration_ms: 2254.0277 type: 'test' ... # Subtest: unfixed dependency URLs and invalid integrity are reported without exposing URL credentials or allowing a plan ok 28 - unfixed dependency URLs and invalid integrity are reported without exposing URL credentials or allowing a plan --- duration_ms: 563.1105 type: 'test' ... # Subtest: hot upgrade refuses an active dependency operation or evaluation before changing runtime objects ok 29 - hot upgrade refuses an active dependency operation or evaluation before changing runtime objects --- duration_ms: 0.5054 type: 'test' ... # Subtest: read-only environment diagnosis identifies actual runtime, missing dependencies and sealed inputs without installing or running scripts ok 30 - read-only environment diagnosis identifies actual runtime, missing dependencies and sealed inputs without installing or running scripts --- duration_ms: 8254.8312 type: 'test' ... # Subtest: diagnosis rejects unsupported managers, lock integrity and credential configuration without reading config content ok 31 - diagnosis rejects unsupported managers, lock integrity and credential configuration without reading config content --- duration_ms: 6565.9403 type: 'test' ... # Subtest: approved real npm ci runs only in owned worktree, skips scripts, retains logs and binds later actual tests ok 32 - approved real npm ci runs only in owned worktree, skips scripts, retains logs and binds later actual tests --- duration_ms: 15619.6044 type: 'test' ... # Subtest: explicitly allowing lifecycle scripts executes sealed root script and preserves the separate preparation result ok 33 - explicitly allowing lifecycle scripts executes sealed root script and preserves the separate preparation result --- duration_ms: 10559.9989 type: 'test' ... # Subtest: stale lockfile, tampered plan/config, cross-session identities and mismatched download sources never dispatch npm ok 34 - stale lockfile, tampered plan/config, cross-session identities and mismatched download sources never dispatch npm --- duration_ms: 8960.0678 type: 'test' ... # Subtest: registry failure keeps original npm exit/log and performs no automatic fallback or replay ok 35 - registry failure keeps original npm exit/log and performs no automatic fallback or replay --- duration_ms: 11295.7391 type: 'test' ... # Subtest: stopping a hanging dependency download records only its owned process and preserves stopped logs ok 36 - stopping a hanging dependency download records only its owned process and preserves stopped logs --- duration_ms: 11862.696 type: 'test' ... # Subtest: restart retains unknown preparation, blocks writes and checks, and requires explicit read-only reconciliation ok 37 - restart retains unknown preparation, blocks writes and checks, and requires explicit read-only reconciliation --- duration_ms: 9217.552 type: 'test' ... # Subtest: HTTP environment flow survives refresh, requires explicit approval and preserves task and artifact ownership ok 38 - HTTP environment flow survives refresh, requires explicit approval and preserves task and artifact ownership --- duration_ms: 15042.2962 type: 'test' ... # Subtest: Python environment diagnosis reads declarations and runtime identity without executing project code or installing packages ok 39 - Python environment diagnosis reads declarations and runtime identity without executing project code or installing packages --- duration_ms: 719.5186 type: 'test' ... # Subtest: empty build cache directories keep dependency identity while added executable files invalidate it ok 40 - empty build cache directories keep dependency identity while added executable files invalidate it --- duration_ms: 954.0431 type: 'test' ... # Subtest: dependency snapshots reject external links and incomplete budgets instead of claiming environment equivalence ok 41 - dependency snapshots reject external links and incomplete budgets instead of claiming environment equivalence --- duration_ms: 594.7706 type: 'test' ... # Subtest: model-requested environment preparation uses the original approval boundary and never treats preparation as passed engineering checks ok 42 - model-requested environment preparation uses the original approval boundary and never treats preparation as passed engineering checks --- duration_ms: 13980.3656 type: 'test' ... # Subtest: timeout and offline cache miss retain actual failures and new attempts require newly sealed plans ok 43 - timeout and offline cache miss retain actual failures and new attempts require newly sealed plans --- duration_ms: 15651.274 type: 'test' ... # Subtest: both evaluation worktrees can approve fixed dependency preparation and different actual dependencies reject candidate execution ok 44 - both evaluation worktrees can approve fixed dependency preparation and different actual dependencies reject candidate execution --- duration_ms: 34041.5956 type: 'test' ... # Subtest: evaluation denominator distinguishes attempted unsealed candidates from unexecuted samples ok 45 - evaluation denominator distinguishes attempted unsealed candidates from unexecuted samples --- duration_ms: 0.3257 type: 'test' ... # Subtest: evaluation summary retains sealed outcomes and stopped or failed original attempts ok 46 - evaluation summary retains sealed outcomes and stopped or failed original attempts --- duration_ms: 0.465 type: 'test' ... # Subtest: evaluation contract fixes bounded samples, source paths, checks and aggregate budget ok 47 - evaluation contract fixes bounded samples, source paths, checks and aggregate budget --- duration_ms: 956.9523 type: 'test' ... # Subtest: same-source workspaces seal real commands before execution and preserve original staged and unstaged bytes ok 48 - same-source workspaces seal real commands before execution and preserve original staged and unstaged bytes --- duration_ms: 17205.081 type: 'test' ... # Subtest: actual original failures and candidate success are both preserved, then adoption requires parent review, checks and Git ok 49 - actual original failures and candidate success are both preserved, then adoption requires parent review, checks and Git --- duration_ms: 61126.989 type: 'test' ... # Subtest: candidate cannot shrink tests and changing original test bytes rejects execution before any model request ok 50 - candidate cannot shrink tests and changing original test bytes rejects execution before any model request --- duration_ms: 22602.9593 type: 'test' ... # Subtest: missing upstream usage remains unknown and does not become a free run or invented cost ok 51 - missing upstream usage remains unknown and does not become a free run or invented cost --- duration_ms: 38489.2214 type: 'test' ... # Subtest: missing raw check logs preserve unknown result and original sample denominator ok 52 - missing raw check logs preserve unknown result and original sample denominator --- duration_ms: 37649.5209 type: 'test' ... # Subtest: source and model changes never silently replace an existing batch input ok 53 - source and model changes never silently replace an existing batch input --- duration_ms: 2746.9231 type: 'test' ... # Subtest: model revision change before preparing a fixed sample cannot silently use the replacement connection ok 54 - model revision change before preparing a fixed sample cannot silently use the replacement connection --- duration_ms: 2777.9777 type: 'test' ... # Subtest: HTTP evaluations require host token, original fingerprint and session ownership; worker API cannot bypass evaluation flow ok 55 - HTTP evaluations require host token, original fingerprint and session ownership; worker API cannot bypass evaluation flow --- duration_ms: 17495.1479 type: 'test' ... # Subtest: sample queue cannot skip or reset attempted samples, and reports freeze further writes ok 56 - sample queue cannot skip or reset attempted samples, and reports freeze further writes --- duration_ms: 31847.5946 type: 'test' ... # Subtest: stop cancels only current candidate and preserves partial results instead of replaying a request ok 57 - stop cancels only current candidate and preserves partial results instead of replaying a request --- duration_ms: 24419.4101 type: 'test' ... # Subtest: restart recovers interrupted evaluations as unknown and keeps completed checks and plan identity ok 58 - restart recovers interrupted evaluations as unknown and keeps completed checks and plan identity --- duration_ms: 21261.4848 type: 'test' ... # Subtest: JSON migration survives process exit before commit and retains exact legacy backups ok 59 - JSON migration survives process exit before commit and retains exact legacy backups --- duration_ms: 1990.0193 type: 'test' ... # Subtest: process exit after_state rolls back state, event and outbox together ok 60 - process exit after_state rolls back state, event and outbox together --- duration_ms: 1599.2708 type: 'test' ... # Subtest: process exit after_event rolls back state, event and outbox together ok 61 - process exit after_event rolls back state, event and outbox together --- duration_ms: 1574.9221 type: 'test' ... # Subtest: crash recovery at accepted retains execution identity and never replays side effects ok 62 - crash recovery at accepted retains execution identity and never replays side effects --- duration_ms: 1590.9582 type: 'test' ... # Subtest: crash recovery at queued retains execution identity and never replays side effects ok 63 - crash recovery at queued retains execution identity and never replays side effects --- duration_ms: 1925.0511 type: 'test' ... # Subtest: crash recovery at approval_wait retains execution identity and never replays side effects ok 64 - crash recovery at approval_wait retains execution identity and never replays side effects --- duration_ms: 2107.0454 type: 'test' ... # Subtest: crash recovery at approved retains execution identity and never replays side effects ok 65 - crash recovery at approved retains execution identity and never replays side effects --- duration_ms: 2315.059 type: 'test' ... # Subtest: crash recovery at effect_unrecorded retains execution identity and never replays side effects ok 66 - crash recovery at effect_unrecorded retains execution identity and never replays side effects --- duration_ms: 2386.9697 type: 'test' ... # Subtest: crash recovery at result_committed retains execution identity and never replays side effects ok 67 - crash recovery at result_committed retains execution identity and never replays side effects --- duration_ms: 2591.4914 type: 'test' ... # Subtest: crash recovery at before_next_model retains execution identity and never replays side effects ok 68 - crash recovery at before_next_model retains execution identity and never replays side effects --- duration_ms: 2593.3901 type: 'test' ... # Subtest: state and event failures publish no success and preserve the last committed memory snapshot ok 69 - state and event failures publish no success and preserve the last committed memory snapshot --- duration_ms: 548.4026 type: 'test' ... # Subtest: tool result artifacts and event cursors retain project ownership across restart ok 70 - tool result artifacts and event cursors retain project ownership across restart --- duration_ms: 1258.1597 type: 'test' ... # Subtest: token budgets pause before tools execute and continuation keeps original task and accumulated usage ok 71 - token budgets pause before tools execute and continuation keeps original task and accumulated usage --- duration_ms: 1346.1656 type: 'test' ... # Subtest: a time budget interrupts an unresponsive model and persists a resumable run ok 72 - a time budget interrupts an unresponsive model and persists a resumable run --- duration_ms: 954.5746 type: 'test' ... # Subtest: a corrupted large result artifact cannot become a recovered success ok 73 - a corrupted large result artifact cannot become a recovered success --- duration_ms: 810.6031 type: 'test' ... # Subtest: hard-link aliases cannot expose protected bytes through reads, proposals, inventory or automatic project rules ok 74 - hard-link aliases cannot expose protected bytes through reads, proposals, inventory or automatic project rules --- duration_ms: 583.9174 type: 'test' ... # Subtest: ambiguous project hard links are explicit and ordinary independent file copies remain readable ok 75 - ambiguous project hard links are explicit and ordinary independent file copies remain readable --- duration_ms: 527.8862 type: 'test' ... # Subtest: proposals preserve project bytes until review; apply backs up the exact original bytes ok 76 - proposals preserve project bytes until review; apply backs up the exact original bytes --- duration_ms: 578.0364 type: 'test' ... # Subtest: new nested files are created only on apply and unchanged content creates no proposal ok 77 - new nested files are created only on apply and unchanged content creates no proposal --- duration_ms: 491.5592 type: 'test' ... # Subtest: stale reads and changed baselines are rejected without overwriting external edits ok 78 - stale reads and changed baselines are rejected without overwriting external edits --- duration_ms: 858.2849 type: 'test' ... # Subtest: file paths reject traversal, absolute paths, credentials and outside junctions ok 79 - file paths reject traversal, absolute paths, credentials and outside junctions --- duration_ms: 807.9712 type: 'test' ... # Subtest: read and search expose bounded text scopes and reject non UTF-8 / oversized files ok 80 - read and search expose bounded text scopes and reject non UTF-8 / oversized files --- duration_ms: 783.6611 type: 'test' ... # Subtest: project-local junctions cannot expose or create protected Git and credential files ok 81 - project-local junctions cannot expose or create protected Git and credential files --- duration_ms: 576.1991 type: 'test' ... # Subtest: Windows trailing-dot and trailing-space path aliases cannot bypass protected paths ok 82 - Windows trailing-dot and trailing-space path aliases cannot bypass protected paths --- duration_ms: 721.4413 type: 'test' ... 1..82 # tests 82 # suites 0 # pass 82 # fail 0 # cancelled 0 # skipped 0 # todo 0 # duration_ms 479922.733