TAP version 13 # Subtest: engine detection seals the actual CLI, SDK, profile and Node and rejects mixed versions without launch ok 1 - engine detection seals the actual CLI, SDK, profile and Node and rejects mixed versions without launch --- duration_ms: 1575.0908 type: 'test' ... # Subtest: owned engine probe waits for real ready identity, verifies six native responses and saves immutable task evidence ok 2 - owned engine probe waits for real ready identity, verifies six native responses and saves immutable task evidence --- duration_ms: 2403.7331 type: 'test' ... # Subtest: an invalid startup nonce is rejected before initialize and preserves the process close and partial report ok 3 - an invalid startup nonce is rejected before initialize and preserves the process close and partial report --- duration_ms: 2269.7652 type: 'test' ... # Subtest: SDK identity and session directory mismatches retain raw responses and never become verified ok 4 - SDK identity and session directory mismatches retain raw responses and never become verified --- duration_ms: 6836.5707 type: 'test' ... # Subtest: cancel during a pending initialize closes only the owned process without sending an unsupported MCP cancellation ok 5 - cancel during a pending initialize closes only the owned process without sending an unsupported MCP cancellation --- duration_ms: 2474.5447 type: 'test' ... # Subtest: cancel before readiness sends no protocol request and a user retry creates a different process and record ok 6 - cancel before readiness sends no protocol request and a user retry creates a different process and record --- duration_ms: 3102.0062 type: 'test' ... # Subtest: source or scratch changes and unexpected model traffic fail after preserving the actual observed result ok 7 - source or scratch changes and unexpected model traffic fail after preserving the actual observed result --- duration_ms: 7625.8371 type: 'test' ... # Subtest: unowned idle notifications stay quarantined and server requests receive a rejection without granting operations ok 8 - unowned idle notifications stay quarantined and server requests receive a rejection without granting operations --- duration_ms: 5275.9709 type: 'test' ... # Subtest: engine probe environment excludes parent secrets and NODE_OPTIONS while preserving operating system paths ok 9 - engine probe environment excludes parent secrets and NODE_OPTIONS while preserving operating system paths --- duration_ms: 2474.7887 type: 'test' ... # Subtest: foreign sessions and altered reports or task bindings cannot reuse verified engine evidence ok 10 - foreign sessions and altered reports or task bindings cannot reuse verified engine evidence --- duration_ms: 2379.8901 type: 'test' ... # Subtest: restart never replays an active engine check or stops a historical PID, and keeps prior task state ok 11 - restart never replays an active engine check or stops a historical PID, and keeps prior task state --- duration_ms: 3117.7888 type: 'test' ... # Subtest: model argument validation and runtime activity do not grant external engine prompt or project execution ok 12 - model argument validation and runtime activity do not grant external engine prompt or project execution --- duration_ms: 1478.732 type: 'test' ... # Subtest: a follow-up question receives the actual failed engine check from its original Task and keeps its report unchanged ok 13 - a follow-up question receives the actual failed engine check from its original Task and keeps its report unchanged --- duration_ms: 4507.3985 type: 'test' ... # Subtest: deep Windows data directories keep REA output within the external path allowance and retain owned evidence ok 14 - deep Windows data directories keep REA output within the external path allowance and retain owned evidence --- duration_ms: 4123.435 type: 'test' ... # Subtest: user research registration creates an owned Task without model requests and seals the exact artifact identity ok 15 - user research registration creates an owned Task without model requests and seals the exact artifact identity --- duration_ms: 1473.7135 type: 'test' ... # Subtest: complete research binds original Evidence and Bundle, pages exact fields and reuses verified results after reload ok 16 - complete research binds original Evidence and Bundle, pages exact fields and reuses verified results after reload --- duration_ms: 3060.4024 type: 'test' ... # Subtest: artifact changes before analysis refuse dispatch and changes during analysis preserve unverified original responses ok 17 - artifact changes before analysis refuse dispatch and changes during analysis preserve unverified original responses --- duration_ms: 4703.5198 type: 'test' ... # Subtest: owned cancellation closes the connection, concurrent and automatic retries are refused, explicit new analysis preserves history ok 18 - owned cancellation closes the connection, concurrent and automatic retries are refused, explicit new analysis preserves history --- duration_ms: 3300.4813 type: 'test' ... # Subtest: a dispatched transport timeout stays unknown and reload never launches or kills a saved PID ok 19 - a dispatched transport timeout stays unknown and reload never launches or kills a saved PID --- duration_ms: 2251.5533 type: 'test' ... # Subtest: corrupt Evidence parts and changed binding records are refused instead of silently re-analysed ok 20 - corrupt Evidence parts and changed binding records are refused instead of silently re-analysed --- duration_ms: 3096.7359 type: 'test' ... # Subtest: research documents over the single-artifact limit preserve exact Unicode bytes with owned verifiable parts ok 21 - research documents over the single-artifact limit preserve exact Unicode bytes with owned verifiable parts --- duration_ms: 1585.3958 type: 'test' ... # Subtest: research tools are only sent for registered Task artifacts and actual HTTP uses the sealed tool contract without replay not ok 22 - research tools are only sent for registered Task artifacts and actual HTTP uses the sealed tool contract without replay --- duration_ms: 12145.9508 type: 'test' location: 'D:\\Ai-project\\REA逆向\\Agent-Workbench\\tests\\research.test.mjs:165:1' failureType: 'testCodeFailure' error: |- Expected values to be strictly equal: 6 !== 5 code: 'ERR_ASSERTION' name: 'AssertionError' expected: 5 actual: 6 operator: 'strictEqual' stack: |- TestContext. (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/tests/research.test.mjs:191:10) runNextTicks (node:internal/process/task_queues:65:5) process.processTimers (node:internal/timers:615:9) async Test.run (node:internal/test_runner/test:1409:7) async Test.processPendingSubtests (node:internal/test_runner/test:974:7) ... # Subtest: file identity rejects directory and invalid formats without running a target ok 23 - file identity rejects directory and invalid formats without running a target --- duration_ms: 1333.5165 type: 'test' ... # Subtest: operational questions are diagnostic while explicit repair requests remain execution instructions ok 24 - operational questions are diagnostic while explicit repair requests remain execution instructions --- duration_ms: 1.0044 type: 'test' ... # Subtest: an unsubstantiated text claim cannot complete a requested file delivery ok 25 - an unsubstantiated text claim cannot complete a requested file delivery --- duration_ms: 2434.0593 type: 'test' ... # Subtest: HTML without tests ends at reviewed file delivery and never offers automatic rechecking ok 26 - HTML without tests ends at reviewed file delivery and never offers automatic rechecking --- duration_ms: 4810.8565 type: 'test' ... # Subtest: diagnostic answers retain the unfinished task, limits and check result, and enforce read-only tools ok 27 - diagnostic answers retain the unfinished task, limits and check result, and enforce read-only tools --- duration_ms: 4850.4281 type: 'test' ... # Subtest: review continuation is a system action and adds no testing demand or fake user instruction ok 28 - review continuation is a system action and adds no testing demand or fake user instruction --- duration_ms: 3447.776 type: 'test' ... # Subtest: a provider cannot execute tests against an explicit no-test task even if it ignores advertised tools ok 29 - a provider cannot execute tests against an explicit no-test task even if it ignores advertised tools --- duration_ms: 2902.5252 type: 'test' ... # Subtest: a budget pause becomes visible only after its input snapshot can be sealed with the original Run ok 30 - a budget pause becomes visible only after its input snapshot can be sealed with the original Run --- duration_ms: 2392.0384 type: 'test' ... # Subtest: alternating duplicate file reads and listing files still pauses without treating new pages as duplicates ok 31 - alternating duplicate file reads and listing files still pauses without treating new pages as duplicates --- duration_ms: 20758.066 type: 'test' ... 1..31 # tests 31 # suites 0 # pass 30 # fail 1 # cancelled 0 # skipped 0 # todo 0 # duration_ms 79011.9795