114 lines
11 KiB
Plaintext
114 lines
11 KiB
Plaintext
✔ reviewed application invalidates old checks, names current SHA and advances through a new approved check to Git preview (7284.684ms)
|
||
✔ skip-tests, foreign task, unknown operations, diagnostic and parent review retain their boundaries (3218.4222ms)
|
||
✔ before the first change, real baseline failures persist across later reads; passed baselines and changed sources cannot certify delivery (7160.4744ms)
|
||
✔ routine engineering facts remove duplicate success excerpts while complete diagnostics and errors remain available (712.9723ms)
|
||
✔ a question about the previous workbench error receives local error context and current change states (1798.8529ms)
|
||
✔ an unknown model request stops, and explicit resume never replays the approved command (2495.5897ms)
|
||
✔ exhausted refused connections preserve tool results, and explicit resume adds no new user turn or replay (2118.0037ms)
|
||
✔ stopping during model backoff cancels further requests and clears the running request state (1411.9489ms)
|
||
✔ HTTP authentication errors are classified, redacted and never automatically retried (17.9515ms)
|
||
✔ temporary HTTP service errors retry and recover with recorded attempt counts (20.8087ms)
|
||
✔ invalid and oversized protocol responses fail once with a useful response-format diagnosis (4.8117ms)
|
||
✔ HTTP rate limits expose their source and obey the service retry interval before recovery (1006.6042ms)
|
||
✔ Windows commands preserve Chinese through a PowerShell pipeline into Node (631.3626ms)
|
||
✔ Windows shell syntax errors stay readable and native command exit codes are retained (1090.6304ms)
|
||
✔ review and completed states do not turn a generic continue click into a new model request (769.0763ms)
|
||
✔ old session data receives a truthful historical connection diagnosis without rewriting history (557.1476ms)
|
||
✔ sealed r1-off failure fits its original 16K config with authentic latest status and readable source (12.9877ms)
|
||
✔ sealed r1-on failure fits its original 16K config with authentic latest status and readable source (9.785ms)
|
||
✔ sealed r2-on failure fits its original 16K config with authentic latest status and readable source (7.5316ms)
|
||
✔ sealed r2-off failure fits its original 16K config with authentic latest status and readable source (7.0282ms)
|
||
✔ archived native call pairs require exact saved arguments/results rather than model-written summaries (8.7903ms)
|
||
✔ full system/user requirements and mandatory current failure can refuse rather than be trimmed (11.9071ms)
|
||
✔ small native input is byte-preserving; developer instructions remain mandatory after compaction (0.1657ms)
|
||
✔ older failing checks and pending proposals remain visible after a later large file read (14.4164ms)
|
||
✔ bounded result keeps every check status and cuts Unicode only on whole characters (0.4407ms)
|
||
✔ continuation reads advance through historical pages without a false pause while repeated current-Run pages still pause (1.9157ms)
|
||
✔ a new tool page replaces old latest-feedback space and still obeys its escaped protocol budget (0.4401ms)
|
||
✔ budgeted artifact pages count escaped protocol bytes and reconstruct original multilingual JSON without losing characters (9.0476ms)
|
||
✔ known artifact coverage detects semantic repeats despite changing limits, and advancing pages or direct evidence restore progress (0.352ms)
|
||
✔ actual large artifact reads are present in the next bounded HTTP request and next_offset progresses without redoing tools (3213.2698ms)
|
||
✔ a real repeated artifact loop pauses the same Task after recorded results, survives reload and only resumes explicitly (5474.5038ms)
|
||
✔ actual HTTP continuation after reload sends distinct historically covered pages without false no_progress or replay (6483.9596ms)
|
||
✔ continuations distinguish unread source, later original line ranges and already returned result wrappers (638.5985ms)
|
||
✔ pure-file batch returns unused small-page space while mixed and historical allocation stays conservative (0.3198ms)
|
||
✔ observed upstream high input pauses before a new request, preserves reported usage and distinguishes the limiting ledger (629.2444ms)
|
||
✔ chat-completions: complete actual Unicode/CRLF pages and small source survive the next bounded wire request (3452.7194ms)
|
||
✔ responses: complete actual Unicode/CRLF pages and small source survive the next bounded wire request (3265.5433ms)
|
||
✔ anthropic-messages: complete actual Unicode/CRLF pages and small source survive the next bounded wire request (3291.3409ms)
|
||
✔ context APIs retain exact request identity, paginate archived groups and reject foreign sessions/runs (4112.6712ms)
|
||
✔ delivery reports verify request body, sealed plan and original archived bytes, and show corruption as needing verification (2447.7783ms)
|
||
✔ native seed framing reserves capacity before optional history without changing the selected window (979.7585ms)
|
||
✔ direct code evidence remains selected while recent historical-index browsing is archived within the complete request budget (1403.987ms)
|
||
✔ one long user turn becomes bounded complete groups with exact UTF-8 history and recent JSON arguments (3953.0325ms)
|
||
✔ tool definitions, exact system bytes, output reserve and remaining Run/Task budget are accounted before requests (818.0986ms)
|
||
✔ oversized required user instructions pause intact and can be explicitly resumed after increasing the window (1595.8903ms)
|
||
✔ all task instructions, constraints and acceptance remain in the runtime system instead of an eight-item or byte prefix (992.4166ms)
|
||
✔ a newest manageable multi-tool group stays exact even beyond the working target; oversized earlier groups are archived (843.6233ms)
|
||
✔ invalid orphan result is rejected before archive or network preparation (577.5376ms)
|
||
✔ invalid missing result is rejected before archive or network preparation (552.2745ms)
|
||
✔ invalid wrong result is rejected before archive or network preparation (528.6487ms)
|
||
✔ invalid duplicate call is rejected before archive or network preparation (707.636ms)
|
||
✔ a group larger than the artifact limit uses verified Unicode-safe chunks without replacing tool JSON (1990.2654ms)
|
||
✔ changed or missing archived evidence stops reuse and is never silently recreated (750.5565ms)
|
||
✔ transaction failure prevents publishing a plan and any corresponding model request (803.6148ms)
|
||
✔ the Provider rejects body mutation after a sealed plan before calling HTTP (686.4617ms)
|
||
✔ HTTP failure and explicit same-Task resume reuse exact completed results and request hashes without repeating a command (8326.1142ms)
|
||
✔ a real process exit after plan persistence marks interruption and never issues an automatic model request (1430.2479ms)
|
||
✖ a real two-reply failure replay reaches request three at 16K with complete goal, original sources and sealed tool evidence (8241.1699ms)
|
||
✔ request deduplication retains long original intent, extra requirements, constraints and acceptance without changing the stored contract (1805.2576ms)
|
||
✔ only the routine context avoids plan echoes; diagnostic API and requests without an included plan keep original observations (591.1168ms)
|
||
✔ the latest oversized session result reaches the next request together with already read source modules (2105.1189ms)
|
||
✔ every call in a large latest group keeps its actual outcome, including denied and unknown results (1473.6871ms)
|
||
✔ a small complete latest tool group is sent once and needs no duplicate feedback view (1091.0731ms)
|
||
✔ real HTTP repeated state inspection pauses after three identical engineering outcomes and preserves tool pairing (4195.8245ms)
|
||
✔ all actually returned same-version fragments reach the bounded request, including overlap and Unicode, with independent original provenance (4300.9099ms)
|
||
✔ unread gaps and different range origins remain separate; partial sending names the precise missing already-observed range (927.5518ms)
|
||
✔ same-SHA conflicting overlaps and invalid coordinates fail closed, while a new SHA never inherits old fragments (1022.0439ms)
|
||
✔ versioned old views keep the original single-page semantics and verification (731.3504ms)
|
||
✔ small code and actual failed command stay visible when their original multi-tool group cannot fit (1091.5887ms)
|
||
✔ mixed group retains six distinct small file observations with exact multilingual text and bounded request bytes (1540.3818ms)
|
||
✔ working observations are bound to the current Task and stop treating a pre-application file revision as current evidence (1569.4472ms)
|
||
✔ a sealed observation rejects changed or missing original result artifacts instead of recreating evidence (1030.1386ms)
|
||
✔ a real passed-check observation is excluded after a source edit invalidates the check, while its historical sent evidence remains valid (4062.1428ms)
|
||
✔ budgeted file pages reconstruct UTF-8 text including long single lines and surrogate boundaries (4.9987ms)
|
||
✔ actual HTTP multi-tool reads survive an oversized runtime snapshot and the sealed evidence survives reopening (8150.4621ms)
|
||
✔ actual HTTP file pagination advances over a long file with each returned page visible in the next request (5260.514ms)
|
||
✔ later mixed history pages cannot displace the actual module sources and complete requirements (4840.5539ms)
|
||
✔ a later narrow page of the same file revision cannot evict the wider actual observation (3012.1338ms)
|
||
✔ six small source files stay complete across metadata turns even when mandatory evidence exceeds the soft target (3093.7698ms)
|
||
✔ historical read-artifact navigation points to the original artifact and next page, never its result wrapper (960.0703ms)
|
||
✔ a 1.6.1 full-provenance sent view remains verifiable after introducing the compact view (1087.7071ms)
|
||
✔ actual HTTP requests keep four observed modules after a later large history read and preserve all source bytes (9154.7472ms)
|
||
ℹ tests 82
|
||
ℹ suites 0
|
||
ℹ pass 81
|
||
ℹ fail 1
|
||
ℹ cancelled 0
|
||
ℹ skipped 0
|
||
ℹ todo 0
|
||
ℹ duration_ms 106880.438
|
||
|
||
✖ failing tests:
|
||
|
||
test at tests\task-context.test.mjs:19:1
|
||
✖ a real two-reply failure replay reaches request three at 16K with complete goal, original sources and sealed tool evidence (8241.1699ms)
|
||
AssertionError [ERR_ASSERTION]: Expected values to be strictly equal:
|
||
+ actual - expected
|
||
|
||
+ undefined
|
||
- 'export function amountToCents(value) { return 0; }\n'
|
||
|
||
at TestContext.<anonymous> (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/tests/task-context.test.mjs:55:12)
|
||
at runNextTicks (node:internal/process/task_queues:65:5)
|
||
at process.processTimers (node:internal/timers:615:9)
|
||
at async Test.run (node:internal/test_runner/test:1409:7)
|
||
at async startSubtestAfterBootstrap (node:internal/test_runner/harness:387:3) {
|
||
generatedMessage: true,
|
||
code: 'ERR_ASSERTION',
|
||
actual: undefined,
|
||
expected: 'export function amountToCents(value) { return 0; }\n',
|
||
operator: 'strictEqual',
|
||
diff: 'simple'
|
||
}
|