✔ agents: a held role with nothing on the bus is idle; a role session gets the claim rows unchanged (150.890306ms) ✔ agents: working counts what the holder received and has not read; routed alone is not working (199.350773ms) ✔ agents: input needed is the role's open decisions routed to the human, gated first, then oldest (335.868521ms) ✔ agents: working wins over waiting, and Input needed still shows (158.368518ms) ✔ agents: a new holder doesn't inherit what the last one left unread; decisions stay with the role (206.929609ms) ✔ agents: state reads only its own business, even for a run name both use (186.357302ms) ✔ agents: a message the holder released unread is stranded; the next holder gets nothing and cannot read it (234.522298ms) ✔ agents: a message the holder read is not stranded; one the current holder has not read is not listed (213.56268ms) ✔ agents: stranded is oldest first and capped at 50 for a role (903.373985ms) ✔ agents: stranded is per role and per business (310.073427ms) ✔ agents: a message routed to a held role and not yet received is not stranded (118.590516ms) ✔ launch identity is stamped, payload identity is refused and stale holder cannot send (166.06202ms) ✔ decision classes route from policy; gated resolution is human-only, choice and target must match (249.015609ms) ✔ claim exclusion, holder release, gated revoke and rerouting to a new holder are atomic (279.298037ms) ✔ launch events require a human CLI capability; generic emit cannot forge authority events (177.154788ms) ✔ within-role decisions close atomically and invalid options or blocking omissions refuse (185.635754ms) ✔ a recommendation that is not exactly one option key refuses invalid-recommendation after the options (282.826811ms) ✔ observer capabilities read human inbox but cannot mutate or forge launch identity (191.502299ms) ✔ task action subjects and linked decision trail are complete and ordered (208.68307ms) ✔ launch binding is durable and reconnecting requires the identical trusted record (130.310089ms) ✔ business isolation includes inherited object names and cross-business message references (204.436747ms) ✔ authority never transfers between action, run, target, unresolved or replaced role holder (267.80783ms) ✔ task projection uses schema current view, skipping earlier and equal-start polls (174.459332ms) ✔ revocation permanently bars the old run from reclaiming first, including after broker restart (208.010738ms) ✔ empty message references refuse before storage; refusal-evidence failure stays a typed error (137.136622ms) ✔ both arbiters require human resolution when their cross-role route is themselves (218.59623ms) ✔ the shipped coder role raises task.reassign as gated to the human; task.scope.change stays cross-role (143.477456ms) ✔ S1 adapter takes resolved limits and refs, rejects mismatched instance, never mutates input (3.651253ms) ✔ credentials gives human and reader sessions service, account, role, date and state, and nothing else (94.500249ms) ✔ credentials refuses a role session and any argument (172.62452ms) ✔ credentials reads the status on each request, and a broker without one lists none (222.412115ms) ✔ the broker process answers credentials without a token, path or variable in its output, log or store (182.874762ms) ✔ only validated broker references load; returned data and exceptions cannot expose a known token (3.999869ms) ✔ bad file modes, symlinks, repository/data paths, malformed tokens and missing dates refuse (4.511286ms) ✔ expiry refuses use and env references never become client data (0.867036ms) ✔ S1 parsed service refs work, service mismatch refuses, Gitea rotation due is a warning state (1.755848ms) ✔ opaque tokens shorter than 16 characters refuse before use (0.403383ms) ✔ a human resolve sends the raising role one message from the human, and its request is the answer (236.134124ms) ✔ a cross-role decision resolved by its arbiter comes from that role and run, with no request (239.841582ms) ✔ a human withdraw sends one message with the note (231.647799ms) ✔ a human seen before the close leaves the request on the close's answer, not the seen (289.297444ms) ✔ the raising run's withdraw, a supersede, a within-role raise and seen send nothing (295.057823ms) ✔ refused closes send nothing: invalid-choice, decision-closed, not-resolver (311.133535ms) ✔ a raising role nobody holds at the close gets the message on the next claim (277.161096ms) ✔ a human DECISION send keeps its own instruction as the request, not a close's answer (249.548326ms) ✔ roles lists open decisions routed to a role, and never the human inbox (210.962475ms) ✔ recent lists closed decisions of every route, newest close first, with how each closed (281.008836ms) ✔ recent gives the closing event as closed when the decision was seen first (274.654702ms) ✔ recent stops at RECENT rows (840.584832ms) ✔ every row is what trail already gives the same session for that id (277.019431ms) ✔ a role session can't list decisions, so it can't read another role's open decisions (395.005795ms) ✔ decisions is a read: one business, strict arguments, nothing written (398.919889ms) ✔ endLaunch writes session.ended, releases the run claim and kills its capabilities (139.794073ms) ✔ endLaunch refuses an unknown run, a second end and a rebind of the ended run (131.033345ms) ✔ a restarted broker refuses to rebind an ended run; a refused rebind leaves the run unbound (161.199958ms) ✔ endLaunch leaves a claim another run took alone (181.486296ms) ✔ refuse records action.refused against the caller with the code only (126.237302ms) ✔ launches off refuses role.launch with launch-revoked until launches on (189.983039ms) ✔ broker process: launch ops authorize role.launch, record refusals and end runs (270.268004ms) ✔ human proof binds CLI entry, process start and nonce; agents and incomplete ancestry refuse (2.629338ms) ✔ process reader gets own kernel identity without exposing environment values (0.472131ms) ✔ EACCES ancestor environments skip only markers; commands and registered launches still refuse (0.736096ms) ✔ real pid 1 remains inspectable when its environment is protected (0.348474ms) ✔ limits binds actor from human proof and rejects role, reader, tester and forged actor (197.634111ms) ✔ runtime wires verified human settings to the actual shared file writer and actor receipt (273.978023ms) ✔ uncertain organization receipts leave bus startup and reads alone but refuse writes until reconciled (416.217589ms) ✔ within-role sends cite an open gated launch decision without spending it or naming it in grants (182.780519ms) ✔ missing and foreign-business citations refuse and roll back message and grant (188.292209ms) ✔ cross-role sends still need a matching resolved decision and consume it once (250.559613ms) ✔ messages pages all human messages newest first without writes; read state is history (2599.479529ms) ✔ messages refuses role callers, invalid paging and cursors outside this human feed (245.593106ms) ✔ views messages passes only its paging arguments through (0.257316ms) ✔ messages isolates businesses even with a valid foreign cursor (153.734385ms) ✔ the assistant sends to the user and to the chief of staff only (283.983721ms) ✔ an assistant send to any other role, or to the tester, is refused and logged (343.919038ms) ✔ only the user and the chief of staff may send to the assistant; the tester is refused by default (330.560929ms) ✔ a sender-side list alone refuses a send to a role that has no list (321.22946ms) ✔ a target-side list alone refuses a send from a role that has no list (341.371828ms) ✔ the sender list is checked before the target list (283.998629ms) ✔ roles without the field behave exactly as before (399.661498ms) ✔ a business with no role definitions at all is unlimited (303.760646ms) ✔ a malformed messaging field refuses at boot (109.99098ms) ✔ the assistant's review.request to a reviewer is refused and logged, even with an approved decision (276.744006ms) ✔ a review.verdict to the assistant from a role outside its receiveFrom is refused and logged (303.452841ms) ✔ review requests and verdicts between roles without the field are unchanged (221.374285ms) ✔ broker process binds trusted launches, offers reader capabilities, refuses human mutation, closes cleanly (197.611624ms) ✔ startup token refusal returns safe code without value or partial listening broker (42.387018ms) ✔ loaded fixture token is absent from socket replies and SQLite, including refusal evidence (193.511503ms) ✔ killed broker leaves an explicit stale lock; another process cannot silently reclaim it (162.185531ms) ✔ trusted host registers later launches; socket clients never have a registration verb (192.614352ms) ✔ runtime excludes declared project roots even when host supplies no repoRoots (37.937261ms) ✔ a refused launch binding leaves the broker and existing capabilities alive; bad protocol stops it (176.034327ms) ✔ a reply is listed for human and reader sessions, and its delivery rows are unchanged (222.11192ms) ✔ the reply still reaches the human message.receive (mosaic talk) after the Console showed it (196.102393ms) ✔ a message with no reply gives an empty list (163.240241ms) ✔ replies come oldest first and stop at REPLIES (770.260935ms) ✔ an id that names no message in the business refuses message-not-found, and the refusal is recorded (319.063668ms) ✔ strict arguments; a role session gets human-required (391.375947ms) ✔ views passes replies through as one call with the id (0.244539ms) ✔ requirements returns the committed list to a holding role session, and no one else (218.300253ms) ✔ the PM's prd.approve with no task, when no id fits, reaches the human's Inbox (149.685628ms) ✔ the PM's non-blocking role.revoke with no task reaches the human's Inbox (189.997471ms) ✔ a push is recorded as the action.allowed for git.push.working on the task, and only the host records one (252.429621ms) ✔ review.request names the latest pushed commit and sends the reviewer a REVIEW-REQUEST (249.22826ms) ✔ review.request says which run holds the reviewer role, or null, and a null request waits for a claim (289.683998ms) ✔ review.request refuses a commit that isn't the latest push, a non-reviewer and a role without the action (381.422917ms) ✔ reviewFetch answers only the addressed reviewer, for the request it cites (267.39716ms) ✔ review.verdict answers one request once, to the role that asked (308.590312ms) ✔ task.close on a pushed task needs an approve verdict for the latest push (383.096375ms) ✔ a refusal the host decided is filed on the task it names, and only a task ref counts (274.82629ms) ✔ v3b prototype refusals, views and append-only mutations (1068.880573ms) ✔ gated approval authorizes once, survives store reopen, and fresh approval works (246.062459ms) ✔ another run cannot consume an approval; a failed check leaves it usable (228.719767ms) ✔ two scheduled callers have exactly one grant and one consumed refusal (170.561974ms) ✔ failed commit rolls consumption back; cross-role consumes and within-role stays reusable (341.408034ms) ✔ class drift gated to cross-role refuses before consumption (234.750422ms) ✔ class drift cross-role to gated refuses before consumption (263.028265ms) ✔ class drift gated to within-role refuses before consumption (214.426149ms) ✔ class drift cross-role to within-role refuses before consumption (250.466151ms) ✔ class drift within-role to gated refuses before consumption (192.990811ms) ✔ class drift within-role to cross-role refuses before consumption (180.089348ms) ✔ message.send consumes approval and prevents a later send or authorize (188.926664ms) ✔ role.revoke consumes approval and prevents a later revoke or authorize (219.226923ms) ✔ a business that never switched launches reads allowed with no time and no event (127.700116ms) ✔ launches revoked and then restored read allowed, with the restore time and event id (157.874498ms) ✔ a task closed on the tracker leaves the open list, and total counts the open ones (149.650991ms) ✔ a tombstone is not open (151.681582ms) ✔ the open list stops at the cap and says so; total still counts every open task (145.685247ms) ✔ a role and a reader session read state; arguments, and an unbound cap, refuse (150.721556ms) ✔ a human session reads the same state, with the same cap (147.791064ms) ✔ reading state writes nothing: every table keeps its row count (176.705427ms) ✔ creates private WAL store and excludes a second writer until explicit close (129.919352ms) ✔ rollback is atomic and schema metadata is checked against trusted DDL, not just itself (168.796796ms) ✔ existing empty database and symlink runtime directory refuse, never initialize over damage (194.121255ms) ✔ crash during a transaction recovers no partial event after explicit fixture-only lock removal (182.056651ms) ✔ writer refuses mixed at/read_at forms atomically, even through trusted SQL helpers (119.550918ms) ✔ async transactions refuse before invoking their function (98.651849ms) ✔ the session default fits under the reply cap and pages to the end with next (323.400573ms) ✔ the short form: no description, assignees by role, every listed key (272.767493ms) ✔ each filter, alone and together, and total counts every match (325.587523ms) ✔ task_ref reads one task in full, as the whole board has it (282.085733ms) ✔ no arguments is the read the Console and mosaic make, unchanged (279.46167ms) ✔ an unknown argument or a bad value refuses invalid-request (869.805341ms) ✔ tasks rows carry priority and requirement for every session that reads tasks (312.259177ms) ✔ the read model rebuilds from the ledger: a reopened bus and each trail give the same values (361.32517ms) ✔ an empty business and one with snapshots but no task.created read cleanly (113.17458ms) ✔ recordTask keeps sync reads and a role write apart (178.424057ms) ✔ read_at must be one canonical UTC format, so the projection compares strings safely (120.119324ms) ✔ a bad entry refuses the whole record (111.866512ms) ✔ taskView reads the projection for one business (134.631728ms) ✔ requestTask hands only a holder and a task verb to the handler, and records refusals (256.611762ms) ✔ the server sends task verbs to the adapter with its own timeout; other verbs stay synchronous (455.033537ms) ✔ without an adapter the server refuses every task verb (190.831846ms) ✔ the runtime refuses an invalid adapter and closes a valid one (205.331132ms) ✔ the process loads the S3 adapter from plain-data trackers (258.959628ms) ✔ the tester binds only when the business sets bus.tester to exactly true (803.483144ms) ✔ a business may not name a role tester, so the tester address is never a role (255.367266ms) ✔ every verb outside the tester list refuses tester-refused and is recorded against the tester (542.091434ms) ✔ a tester message is from tester, starts [mos test], and writes no human.input or authority (192.989227ms) ✔ a role answers a tester only in reply to a tester message, and the answer never reaches the human (216.091204ms) ✔ the tester reads the human views a reader reads (137.580699ms) ✔ tester proof binds its own CLI, start time and nonce, refuses a managed run and allows an agent (1.179739ms) ✔ neither CLI's nonce proves the other: each proof is bound to its own CLI and nonce variable (0.366462ms) ✔ testerNonce reads only the tester's variable and gives null on any read error (0.208682ms) ✔ an envelope carries exactly one of cap, human and tester; no tester authenticator refuses (78.341489ms) ✔ tester-cli.mjs binds through the runtime when the switch is on, and refuses off or under a run marker (762.128143ms) ✔ socket capability stamps launch identity; shared views use wire, no SQL client (153.413917ms) ✔ two wire claims serialize; a lost reply never automatically retries (188.01779ms) ✔ malformed, oversized and identity-forging envelopes refuse without echoing input (119.131735ms) ✔ client preserves UTF-8 when a response divides a multibyte character (10.558342ms) ✔ committed mutation followed by dropped reply reports unknown and is never retried (126.485799ms) ℹ tests 171 ℹ suites 0 ℹ pass 171 ℹ fail 0 ℹ cancelled 0 ℹ skipped 0 ℹ todo 0 ℹ duration_ms 4322.236069