Files
Agent-Workbench/docs/working-memory-u-evidence/receipts/focused-third.tap
T
2026-11-04 08:39:17 +08:00

62 lines
5.7 KiB
Plaintext
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
✔ reviewed application invalidates old checks, names current SHA and advances through a new approved check to Git preview (7142.3386ms)
✔ skip-tests, foreign task, unknown operations, diagnostic and parent review retain their boundaries (2848.1591ms)
✖ before the first change, real baseline failures persist across later reads; passed baselines and changed sources cannot certify delivery (5088.5983ms)
✔ a question about the previous workbench error receives local error context and current change states (1625.3057ms)
✔ an unknown model request stops, and explicit resume never replays the approved command (2670.744ms)
✔ exhausted refused connections preserve tool results, and explicit resume adds no new user turn or replay (2076.6573ms)
✔ stopping during model backoff cancels further requests and clears the running request state (1232.15ms)
✔ HTTP authentication errors are classified, redacted and never automatically retried (18.2276ms)
✔ temporary HTTP service errors retry and recover with recorded attempt counts (7.4302ms)
✔ invalid and oversized protocol responses fail once with a useful response-format diagnosis (5.4312ms)
✔ HTTP rate limits expose their source and obey the service retry interval before recovery (1013.389ms)
✔ Windows commands preserve Chinese through a PowerShell pipeline into Node (780.9756ms)
✔ Windows shell syntax errors stay readable and native command exit codes are retained (734.556ms)
✔ review and completed states do not turn a generic continue click into a new model request (497.458ms)
✔ old session data receives a truthful historical connection diagnosis without rewriting history (722.6187ms)
✔ the latest oversized session result reaches the next request together with already read source modules (1505.3271ms)
✔ every call in a large latest group keeps its actual outcome, including denied and unknown results (1214.2756ms)
✔ a small complete latest tool group is sent once and needs no duplicate feedback view (1091.5978ms)
✔ real HTTP repeated state inspection pauses after three identical engineering outcomes and preserves tool pairing (3592.788ms)
✔ all actually returned same-version fragments reach the bounded request, including overlap and Unicode, with independent original provenance (4466.257ms)
✔ unread gaps and different range origins remain separate; partial sending names the precise missing already-observed range (776.9782ms)
✔ same-SHA conflicting overlaps and invalid coordinates fail closed, while a new SHA never inherits old fragments (939.5615ms)
✔ versioned old views keep the original single-page semantics and verification (650.005ms)
✔ small code and actual failed command stay visible when their original multi-tool group cannot fit (947.0277ms)
✔ mixed group retains six distinct small file observations with exact multilingual text and bounded request bytes (1511.7982ms)
✔ working observations are bound to the current Task and stop treating a pre-application file revision as current evidence (1644.6622ms)
✔ a sealed observation rejects changed or missing original result artifacts instead of recreating evidence (1129.2701ms)
✔ a real passed-check observation is excluded after a source edit invalidates the check, while its historical sent evidence remains valid (4193.5591ms)
✔ budgeted file pages reconstruct UTF-8 text including long single lines and surrogate boundaries (4.4328ms)
✔ actual HTTP multi-tool reads survive an oversized runtime snapshot and the sealed evidence survives reopening (7936.4707ms)
✔ actual HTTP file pagination advances over a long file with each returned page visible in the next request (4975.0468ms)
✔ later mixed history pages cannot displace the actual module sources and complete requirements (4365.3575ms)
✔ a later narrow page of the same file revision cannot evict the wider actual observation (2370.8145ms)
✔ six small source files stay complete across metadata turns even when mandatory evidence exceeds the soft target (2455.1951ms)
✔ historical read-artifact navigation points to the original artifact and next page, never its result wrapper (1006.5186ms)
✔ a 1.6.1 full-provenance sent view remains verifiable after introducing the compact view (1015.8284ms)
✔ actual HTTP requests keep four observed modules after a later large history read and preserve all source bytes (9118.6557ms)
ℹ tests 37
ℹ suites 0
ℹ pass 36
ℹ fail 1
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 64878.891
✖ failing tests:
test at tests\delivery-progress.test.mjs:39:1
✖ before the first change, real baseline failures persist across later reads; passed baselines and changed sources cannot certify delivery (5088.5983ms)
Error: 执行记录的身份或项目归属无效。
at ownedRecord (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/shared/contracts/execution.mjs:87:11)
at file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/server/store.mjs:173:22
at Array.map (<anonymous>)
at file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/server/store.mjs:172:117
at Array.flatMap (<anonymous>)
at Store.recordsFor (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/server/store.mjs:172:23)
at Store.save (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/server/store.mjs:386:55)
at Store.putRecord (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/server/store.mjs:665:10)
at CheckRunner.plan (file:///D:/Ai-project/REA%E9%80%86%E5%90%91/Agent-Workbench/src/core/check-runner.mjs:98:23)
at process.processTicksAndRejections (node:internal/process/task_queues:104:5)