jason.woltjeandClaude Opus 5.5
dbf72a2a3e
Candidate on 96656dd21: broker split refuses invalid-recommendation after the
option checks, raise_decision text asks for one key and names the yes default,
bus README codes table, broker and tools tests. Mutants m1 (old shared line)
and m2 (order swapped) killed; full gate passed 15:42:03-15:47:36Z.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
157 lines
14 KiB
Plaintext
157 lines
14 KiB
Plaintext
✔ agents: a held role with nothing on the bus is idle; a role session gets the claim rows unchanged (141.34693ms)
|
||
✔ agents: working counts what the holder received and has not read; routed alone is not working (199.423726ms)
|
||
✔ agents: input needed is the role's open decisions routed to the human, gated first, then oldest (331.57252ms)
|
||
✔ agents: working wins over waiting, and Input needed still shows (170.012612ms)
|
||
✔ agents: a new holder doesn't inherit what the last one left unread; decisions stay with the role (213.395779ms)
|
||
✔ agents: state reads only its own business, even for a run name both use (201.048081ms)
|
||
✔ agents: a message the holder released unread is stranded; the next holder gets nothing and cannot read it (205.814405ms)
|
||
✔ agents: a message the holder read is not stranded; one the current holder has not read is not listed (204.572262ms)
|
||
✔ agents: stranded is oldest first and capped at 50 for a role (881.208188ms)
|
||
✔ agents: stranded is per role and per business (307.075602ms)
|
||
✔ agents: a message routed to a held role and not yet received is not stranded (122.860832ms)
|
||
✔ launch identity is stamped, payload identity is refused and stale holder cannot send (170.071079ms)
|
||
✔ decision classes route from policy; gated resolution is human-only, choice and target must match (270.418729ms)
|
||
✔ claim exclusion, holder release, gated revoke and rerouting to a new holder are atomic (280.269408ms)
|
||
✔ launch events require a human CLI capability; generic emit cannot forge authority events (188.941986ms)
|
||
✔ within-role decisions close atomically and invalid options or blocking omissions refuse (181.554015ms)
|
||
✔ a recommendation that is not exactly one option key refuses invalid-recommendation after the options (211.697819ms)
|
||
✔ observer capabilities read human inbox but cannot mutate or forge launch identity (168.497261ms)
|
||
✔ task action subjects and linked decision trail are complete and ordered (190.03021ms)
|
||
✔ launch binding is durable and reconnecting requires the identical trusted record (127.398309ms)
|
||
✔ business isolation includes inherited object names and cross-business message references (192.772806ms)
|
||
✔ authority never transfers between action, run, target, unresolved or replaced role holder (304.043133ms)
|
||
✔ task projection uses schema current view, skipping earlier and equal-start polls (148.300699ms)
|
||
✔ revocation permanently bars the old run from reclaiming first, including after broker restart (215.048126ms)
|
||
✔ empty message references refuse before storage; refusal-evidence failure stays a typed error (126.12583ms)
|
||
✔ both arbiters require human resolution when their cross-role route is themselves (231.063651ms)
|
||
✔ the shipped coder role raises task.reassign as gated to the human; task.scope.change stays cross-role (127.11875ms)
|
||
✔ S1 adapter takes resolved limits and refs, rejects mismatched instance, never mutates input (3.489268ms)
|
||
✔ credentials gives human and reader sessions service, account, role, date and state, and nothing else (104.962191ms)
|
||
✔ credentials refuses a role session and any argument (177.3956ms)
|
||
✔ credentials reads the status on each request, and a broker without one lists none (211.690017ms)
|
||
✔ the broker process answers credentials without a token, path or variable in its output, log or store (184.294741ms)
|
||
✔ only validated broker references load; returned data and exceptions cannot expose a known token (5.076804ms)
|
||
✔ bad file modes, symlinks, repository/data paths, malformed tokens and missing dates refuse (2.246185ms)
|
||
✔ expiry refuses use and env references never become client data (0.794916ms)
|
||
✔ S1 parsed service refs work, service mismatch refuses, Gitea rotation due is a warning state (1.699117ms)
|
||
✔ opaque tokens shorter than 16 characters refuse before use (0.39942ms)
|
||
✔ roles lists open decisions routed to a role, and never the human inbox (221.730393ms)
|
||
✔ recent lists closed decisions of every route, newest close first, with how each closed (285.225884ms)
|
||
✔ recent gives the closing event as closed when the decision was seen first (303.139357ms)
|
||
✔ recent stops at RECENT rows (808.921339ms)
|
||
✔ every row is what trail already gives the same session for that id (269.39356ms)
|
||
✔ a role session can't list decisions, so it can't read another role's open decisions (426.423131ms)
|
||
✔ decisions is a read: one business, strict arguments, nothing written (371.967682ms)
|
||
✔ endLaunch writes session.ended, releases the run claim and kills its capabilities (116.887971ms)
|
||
✔ endLaunch refuses an unknown run, a second end and a rebind of the ended run (138.645737ms)
|
||
✔ a restarted broker refuses to rebind an ended run; a refused rebind leaves the run unbound (158.132474ms)
|
||
✔ endLaunch leaves a claim another run took alone (181.411852ms)
|
||
✔ refuse records action.refused against the caller with the code only (142.366477ms)
|
||
✔ launches off refuses role.launch with launch-revoked until launches on (201.07398ms)
|
||
✔ broker process: launch ops authorize role.launch, record refusals and end runs (274.429231ms)
|
||
✔ human proof binds CLI entry, process start and nonce; agents and incomplete ancestry refuse (2.860455ms)
|
||
✔ process reader gets own kernel identity without exposing environment values (0.743232ms)
|
||
✔ EACCES ancestor environments skip only markers; commands and registered launches still refuse (0.888538ms)
|
||
✔ real pid 1 remains inspectable when its environment is protected (0.309841ms)
|
||
✔ within-role sends cite an open gated launch decision without spending it or naming it in grants (189.443213ms)
|
||
✔ missing and foreign-business citations refuse and roll back message and grant (194.423036ms)
|
||
✔ cross-role sends still need a matching resolved decision and consume it once (253.555189ms)
|
||
✔ messages pages all human messages newest first without writes; read state is history (2577.203033ms)
|
||
✔ messages refuses role callers, invalid paging and cursors outside this human feed (242.907951ms)
|
||
✔ views messages passes only its paging arguments through (0.273037ms)
|
||
✔ messages isolates businesses even with a valid foreign cursor (162.896965ms)
|
||
✔ broker process binds trusted launches, offers reader capabilities, refuses human mutation, closes cleanly (184.956551ms)
|
||
✔ startup token refusal returns safe code without value or partial listening broker (35.771039ms)
|
||
✔ loaded fixture token is absent from socket replies and SQLite, including refusal evidence (175.757787ms)
|
||
✔ killed broker leaves an explicit stale lock; another process cannot silently reclaim it (172.428142ms)
|
||
✔ trusted host registers later launches; socket clients never have a registration verb (183.26112ms)
|
||
✔ runtime excludes declared project roots even when host supplies no repoRoots (37.552032ms)
|
||
✔ a refused launch binding leaves the broker and existing capabilities alive; bad protocol stops it (180.553703ms)
|
||
✔ a reply is listed for human and reader sessions, and its delivery rows are unchanged (220.504403ms)
|
||
✔ the reply still reaches the human message.receive (mosaic talk) after the Console showed it (197.47024ms)
|
||
✔ a message with no reply gives an empty list (156.009382ms)
|
||
✔ replies come oldest first and stop at REPLIES (782.586723ms)
|
||
✔ an id that names no message in the business refuses message-not-found, and the refusal is recorded (282.618122ms)
|
||
✔ strict arguments; a role session gets human-required (380.216373ms)
|
||
✔ views passes replies through as one call with the id (0.271935ms)
|
||
✔ requirements returns the committed list to a holding role session, and no one else (187.712111ms)
|
||
✔ the PM's prd.approve with no task, when no id fits, reaches the human's Inbox (152.326993ms)
|
||
✔ the PM's non-blocking role.revoke with no task reaches the human's Inbox (165.718562ms)
|
||
✔ a push is recorded as the action.allowed for git.push.working on the task, and only the host records one (222.062376ms)
|
||
✔ review.request names the latest pushed commit and sends the reviewer a REVIEW-REQUEST (242.49925ms)
|
||
✔ review.request says which run holds the reviewer role, or null, and a null request waits for a claim (305.67137ms)
|
||
✔ review.request refuses a commit that isn't the latest push, a non-reviewer and a role without the action (395.052661ms)
|
||
✔ reviewFetch answers only the addressed reviewer, for the request it cites (238.151839ms)
|
||
✔ review.verdict answers one request once, to the role that asked (298.294307ms)
|
||
✔ task.close on a pushed task needs an approve verdict for the latest push (391.848007ms)
|
||
✔ a refusal the host decided is filed on the task it names, and only a task ref counts (267.019532ms)
|
||
✔ v3b prototype refusals, views and append-only mutations (1065.207365ms)
|
||
✔ gated approval authorizes once, survives store reopen, and fresh approval works (263.004223ms)
|
||
✔ another run cannot consume an approval; a failed check leaves it usable (224.3379ms)
|
||
✔ two scheduled callers have exactly one grant and one consumed refusal (163.292853ms)
|
||
✔ failed commit rolls consumption back; cross-role consumes and within-role stays reusable (342.769721ms)
|
||
✔ class drift gated to cross-role refuses before consumption (242.601487ms)
|
||
✔ class drift cross-role to gated refuses before consumption (240.701207ms)
|
||
✔ class drift gated to within-role refuses before consumption (205.703227ms)
|
||
✔ class drift cross-role to within-role refuses before consumption (254.64266ms)
|
||
✔ class drift within-role to gated refuses before consumption (232.952259ms)
|
||
✔ class drift within-role to cross-role refuses before consumption (200.139442ms)
|
||
✔ message.send consumes approval and prevents a later send or authorize (227.508023ms)
|
||
✔ role.revoke consumes approval and prevents a later revoke or authorize (235.716371ms)
|
||
✔ a business that never switched launches reads allowed with no time and no event (138.947963ms)
|
||
✔ launches revoked and then restored read allowed, with the restore time and event id (162.296971ms)
|
||
✔ a task closed on the tracker leaves the open list, and total counts the open ones (135.519943ms)
|
||
✔ a tombstone is not open (150.454401ms)
|
||
✔ the open list stops at the cap and says so; total still counts every open task (147.772622ms)
|
||
✔ a role and a reader session read state; arguments, and an unbound cap, refuse (160.739454ms)
|
||
✔ a human session reads the same state, with the same cap (138.931363ms)
|
||
✔ reading state writes nothing: every table keeps its row count (174.465144ms)
|
||
✔ creates private WAL store and excludes a second writer until explicit close (138.865585ms)
|
||
✔ rollback is atomic and schema metadata is checked against trusted DDL, not just itself (181.689508ms)
|
||
✔ existing empty database and symlink runtime directory refuse, never initialize over damage (182.07818ms)
|
||
✔ crash during a transaction recovers no partial event after explicit fixture-only lock removal (171.172005ms)
|
||
✔ writer refuses mixed at/read_at forms atomically, even through trusted SQL helpers (117.077622ms)
|
||
✔ async transactions refuse before invoking their function (104.414458ms)
|
||
✔ the session default fits under the reply cap and pages to the end with next (294.210081ms)
|
||
✔ the short form: no description, assignees by role, every listed key (269.614094ms)
|
||
✔ each filter, alone and together, and total counts every match (305.846584ms)
|
||
✔ task_ref reads one task in full, as the whole board has it (284.91637ms)
|
||
✔ no arguments is the read the Console and mosaic make, unchanged (249.028492ms)
|
||
✔ an unknown argument or a bad value refuses invalid-request (937.991527ms)
|
||
✔ tasks rows carry priority and requirement for every session that reads tasks (325.217891ms)
|
||
✔ the read model rebuilds from the ledger: a reopened bus and each trail give the same values (347.509596ms)
|
||
✔ an empty business and one with snapshots but no task.created read cleanly (119.882637ms)
|
||
✔ recordTask keeps sync reads and a role write apart (182.138797ms)
|
||
✔ read_at must be one canonical UTC format, so the projection compares strings safely (104.218136ms)
|
||
✔ a bad entry refuses the whole record (119.618367ms)
|
||
✔ taskView reads the projection for one business (122.088791ms)
|
||
✔ requestTask hands only a holder and a task verb to the handler, and records refusals (235.545947ms)
|
||
✔ the server sends task verbs to the adapter with its own timeout; other verbs stay synchronous (406.220158ms)
|
||
✔ without an adapter the server refuses every task verb (242.018924ms)
|
||
✔ the runtime refuses an invalid adapter and closes a valid one (221.222001ms)
|
||
✔ the process loads the S3 adapter from plain-data trackers (278.175372ms)
|
||
✔ the tester binds only when the business sets bus.tester to exactly true (785.529121ms)
|
||
✔ a business may not name a role tester, so the tester address is never a role (252.469284ms)
|
||
✔ every verb outside the tester list refuses tester-refused and is recorded against the tester (573.216055ms)
|
||
✔ a tester message is from tester, starts [mos test], and writes no human.input or authority (211.542237ms)
|
||
✔ a role answers a tester only in reply to a tester message, and the answer never reaches the human (243.987926ms)
|
||
✔ the tester reads the human views a reader reads (113.133958ms)
|
||
✔ tester proof binds its own CLI, start time and nonce, refuses a managed run and allows an agent (1.162779ms)
|
||
✔ neither CLI's nonce proves the other: each proof is bound to its own CLI and nonce variable (0.361729ms)
|
||
✔ testerNonce reads only the tester's variable and gives null on any read error (0.223485ms)
|
||
✔ an envelope carries exactly one of cap, human and tester; no tester authenticator refuses (47.739021ms)
|
||
✔ tester-cli.mjs binds through the runtime when the switch is on, and refuses off or under a run marker (713.775638ms)
|
||
✔ socket capability stamps launch identity; shared views use wire, no SQL client (165.14067ms)
|
||
✔ two wire claims serialize; a lost reply never automatically retries (169.902674ms)
|
||
✔ malformed, oversized and identity-forging envelopes refuse without echoing input (109.746166ms)
|
||
✔ client preserves UTF-8 when a response divides a multibyte character (12.374961ms)
|
||
✔ committed mutation followed by dropped reply reports unknown and is never retried (121.802836ms)
|
||
ℹ tests 148
|
||
ℹ suites 0
|
||
ℹ pass 148
|
||
ℹ fail 0
|
||
ℹ cancelled 0
|
||
ℹ skipped 0
|
||
ℹ todo 0
|
||
ℹ duration_ms 4019.083354
|