114 lines
12 KiB
Plaintext
114 lines
12 KiB
Plaintext
✔ a budget pause persists concrete text/tool/history/image breakdown and never dispatches an oversized request (2191.7892ms)
|
||
✔ explicit frozen preview resumes the same Task in a new Run, preserves completed tools and does not change global config or selected model (2139.4491ms)
|
||
✔ insufficient budgets, silently clamped capacity and arbitrary config fields reject without a physical request or mutating task budget (1673.9553ms)
|
||
✔ changed session, changed registry, expiry and altered frozen preview all require a new explicit preview (2318.5709ms)
|
||
✔ preview preserves unknown invocation status and increasing Task allowance is committed only with confirmed new Run (2755.9992ms)
|
||
✔ reviewed application invalidates old checks, names current SHA and advances through a new approved check to Git preview (7154.5899ms)
|
||
✔ skip-tests, foreign task, unknown operations, diagnostic and parent review retain their boundaries (3444.1476ms)
|
||
✔ a question about the previous workbench error receives local error context and current change states (1463.013ms)
|
||
✔ an unknown model request stops, and explicit resume never replays the approved command (3592.7559ms)
|
||
✔ exhausted refused connections preserve tool results, and explicit resume adds no new user turn or replay (2847.7393ms)
|
||
✔ stopping during model backoff cancels further requests and clears the running request state (1409.735ms)
|
||
✔ HTTP authentication errors are classified, redacted and never automatically retried (20.3864ms)
|
||
✔ temporary HTTP service errors retry and recover with recorded attempt counts (7.2043ms)
|
||
✔ invalid and oversized protocol responses fail once with a useful response-format diagnosis (5.8284ms)
|
||
✔ HTTP rate limits expose their source and obey the service retry interval before recovery (1016.0122ms)
|
||
✔ Windows commands preserve Chinese through a PowerShell pipeline into Node (825.8781ms)
|
||
✔ Windows shell syntax errors stay readable and native command exit codes are retained (1021.759ms)
|
||
✔ review and completed states do not turn a generic continue click into a new model request (500.5327ms)
|
||
✔ old session data receives a truthful historical connection diagnosis without rewriting history (699.4177ms)
|
||
✔ sealed r1-off failure fits its original 16K config with authentic latest status and readable source (12.4639ms)
|
||
✔ sealed r1-on failure fits its original 16K config with authentic latest status and readable source (9.5179ms)
|
||
✔ sealed r2-on failure fits its original 16K config with authentic latest status and readable source (7.4422ms)
|
||
✔ sealed r2-off failure fits its original 16K config with authentic latest status and readable source (7.4418ms)
|
||
✔ archived native call pairs require exact saved arguments/results rather than model-written summaries (7.8064ms)
|
||
✔ full system/user requirements and mandatory current failure can refuse rather than be trimmed (11.7338ms)
|
||
✔ small native input is byte-preserving; developer instructions remain mandatory after compaction (0.1682ms)
|
||
✔ older failing checks and pending proposals remain visible after a later large file read (13.7606ms)
|
||
✔ bounded result keeps every check status and cuts Unicode only on whole characters (0.43ms)
|
||
✔ unchanged proposals cannot hide repeated original-log reads or identical session inspections (5686.5182ms)
|
||
✔ reusing a verified proof never resets identical-state detection or creates a new proof (5033.799ms)
|
||
✔ coverage merges adjacent pages, separates changed sources and never claims old Run coverage in a continuation (0.993ms)
|
||
✔ continuation reads advance through historical pages without a false pause while repeated current-Run pages still pause (1.7713ms)
|
||
✔ a new tool page replaces old latest-feedback space and still obeys its escaped protocol budget (0.4226ms)
|
||
✔ budgeted artifact pages count escaped protocol bytes and reconstruct original multilingual JSON without losing characters (9.6886ms)
|
||
✔ known artifact coverage detects semantic repeats despite changing limits, and advancing pages or direct evidence restore progress (0.3384ms)
|
||
✔ actual large artifact reads are present in the next bounded HTTP request and next_offset progresses without redoing tools (4385.924ms)
|
||
✔ a real repeated artifact loop pauses the same Task after recorded results, survives reload and only resumes explicitly (5535.8999ms)
|
||
✔ actual HTTP continuation after reload sends distinct historically covered pages without false no_progress or replay (6173.3485ms)
|
||
✔ context APIs retain exact request identity, paginate archived groups and reject foreign sessions/runs (3377.2936ms)
|
||
✔ delivery reports verify request body, sealed plan and original archived bytes, and show corruption as needing verification (2226.0169ms)
|
||
✔ native seed framing reserves capacity before optional history without changing the selected window (1001.789ms)
|
||
✔ direct code evidence remains selected while recent historical-index browsing is archived within the complete request budget (2030.2967ms)
|
||
✔ one long user turn becomes bounded complete groups with exact UTF-8 history and recent JSON arguments (3705.7592ms)
|
||
✔ tool definitions, exact system bytes, output reserve and remaining Run/Task budget are accounted before requests (734.5634ms)
|
||
✔ oversized required user instructions pause intact and can be explicitly resumed after increasing the window (1460.3881ms)
|
||
✔ all task instructions, constraints and acceptance remain in the runtime system instead of an eight-item or byte prefix (1342.2269ms)
|
||
✔ a newest manageable multi-tool group stays exact even beyond the working target; oversized earlier groups are archived (949.0773ms)
|
||
✔ invalid orphan result is rejected before archive or network preparation (704.0679ms)
|
||
✔ invalid missing result is rejected before archive or network preparation (692.6761ms)
|
||
✔ invalid wrong result is rejected before archive or network preparation (518.9571ms)
|
||
✔ invalid duplicate call is rejected before archive or network preparation (607.2725ms)
|
||
✔ a group larger than the artifact limit uses verified Unicode-safe chunks without replacing tool JSON (1823.7428ms)
|
||
✔ changed or missing archived evidence stops reuse and is never silently recreated (865.2527ms)
|
||
✔ transaction failure prevents publishing a plan and any corresponding model request (1015.2601ms)
|
||
✔ the Provider rejects body mutation after a sealed plan before calling HTTP (655.1349ms)
|
||
✔ HTTP failure and explicit same-Task resume reuse exact completed results and request hashes without repeating a command (9105.7548ms)
|
||
✔ a real process exit after plan persistence marks interruption and never issues an automatic model request (1254.472ms)
|
||
✔ a real two-reply failure replay reaches request three at 16K with complete goal, original sources and sealed tool evidence (7814.414ms)
|
||
✔ request deduplication retains long original intent, extra requirements, constraints and acceptance without changing the stored contract (1798.7725ms)
|
||
✔ only the routine context avoids plan echoes; diagnostic API and requests without an included plan keep original observations (709.2961ms)
|
||
✔ operational questions are diagnostic while explicit repair requests remain execution instructions (1.0146ms)
|
||
✔ file delivery requires a positive instruction, not saved evidence or a prohibited write (1.3442ms)
|
||
✔ an unsubstantiated text claim cannot complete a requested file delivery (1289.9605ms)
|
||
✔ HTML without tests ends at reviewed file delivery and never offers automatic rechecking (2790.2241ms)
|
||
✔ diagnostic answers retain the unfinished task, limits and check result, and enforce read-only tools (3391.1027ms)
|
||
✔ review continuation is a system action and adds no testing demand or fake user instruction (2597.8317ms)
|
||
✔ a provider cannot execute tests against an explicit no-test task even if it ignores advertised tools (1353.0843ms)
|
||
✔ a budget pause becomes visible only after its input snapshot can be sealed with the original Run (1226.8333ms)
|
||
✔ alternating duplicate file reads and listing files still pauses without treating new pages as duplicates (12273.6651ms)
|
||
✔ task-plan HTTP mutations bind current Task/Run and retain original evidence through real host restart (5023.8798ms)
|
||
✔ concurrent stale HTTP plan revisions cannot replace conditions or attach proofs to a different plan (2819.8705ms)
|
||
✔ a real escaped long-file tool message keeps its proof identity within a tight request budget (1045.5884ms)
|
||
✔ task plans reject fabricated completion, invalid scopes and cyclic or missing dependencies (0.4806ms)
|
||
✔ sealed plans retain goal, contract hash and original steps through SQLite reopen without executing tools (891.2459ms)
|
||
✔ proofs require original owned reads, honor dependencies and reject altered invocation arguments and bytes (1225.4179ms)
|
||
✔ new user requirements invalidate current proofs and reseal the same original operations in a new revision (1252.4869ms)
|
||
✔ applied change-set evidence binds reviewed current files and invalidates dependent steps after outside edits (1659.0351ms)
|
||
✔ actual npm checks are verified by original counted logs and current source, with failure history retained (7398.17ms)
|
||
✔ real desktop and mobile browser proofs bind required actions and PNGs, rejecting source changes, a later failed plan and changed builds (12152.9825ms)
|
||
✔ Git delivery blocks incomplete plans and changed preview bindings, preserving HEAD and owner index before a fresh verified commit (14185.1651ms)
|
||
✔ the runtime supplies owned read IDs and completes a plan only after original step evidence verification (2873.1178ms)
|
||
✔ premature model completion pauses the original Task and resume preserves the existing plan without replay (2650.5911ms)
|
||
✔ plan operations participate in project activity and orderly shutdown (599.6829ms)
|
||
✔ actual process exit before plan save restores only committed plan evidence without replay (1158.5117ms)
|
||
✔ actual process exit after plan save restores only committed plan evidence without replay (1418.7682ms)
|
||
✔ actual process exit before proof save restores only committed plan evidence without replay (1376.7302ms)
|
||
✔ actual process exit after proof save restores only committed plan evidence without replay (1509.5395ms)
|
||
✔ the latest oversized session result reaches the next request together with already read source modules (1665.8206ms)
|
||
✔ every call in a large latest group keeps its actual outcome, including denied and unknown results (1027.2538ms)
|
||
✔ a small complete latest tool group is sent once and needs no duplicate feedback view (861.9708ms)
|
||
✔ real HTTP repeated state inspection pauses after three identical engineering outcomes and preserves tool pairing (2686.3208ms)
|
||
✔ small code and actual failed command stay visible when their original multi-tool group cannot fit (1233.4842ms)
|
||
✔ mixed group retains six distinct small file observations with exact multilingual text and bounded request bytes (1557.7275ms)
|
||
✔ working observations are bound to the current Task and stop treating a pre-application file revision as current evidence (1430.9522ms)
|
||
✔ a sealed observation rejects changed or missing original result artifacts instead of recreating evidence (988.9508ms)
|
||
✔ a real passed-check observation is excluded after a source edit invalidates the check, while its historical sent evidence remains valid (4391.5405ms)
|
||
✔ budgeted file pages reconstruct UTF-8 text including long single lines and surrogate boundaries (4.2794ms)
|
||
✔ actual HTTP multi-tool reads survive an oversized runtime snapshot and the sealed evidence survives reopening (10124.0668ms)
|
||
✔ actual HTTP file pagination advances over a long file with each returned page visible in the next request (7754.314ms)
|
||
✔ later mixed history pages cannot displace the actual module sources and complete requirements (6522.3736ms)
|
||
✔ a later narrow page of the same file revision cannot evict the wider actual observation (3791.9963ms)
|
||
✔ six small source files stay complete across metadata turns even when mandatory evidence exceeds the soft target (2992.4718ms)
|
||
✔ historical read-artifact navigation points to the original artifact and next page, never its result wrapper (1093.8212ms)
|
||
✔ a 1.6.1 full-provenance sent view remains verifiable after introducing the compact view (1043.2575ms)
|
||
✔ actual HTTP requests keep four observed modules after a later large history read and preserve all source bytes (9438.3497ms)
|
||
ℹ tests 105
|
||
ℹ suites 0
|
||
ℹ pass 105
|
||
ℹ fail 0
|
||
ℹ cancelled 0
|
||
ℹ skipped 0
|
||
ℹ todo 0
|
||
ℹ duration_ms 133189.2766
|