✔ agents: a held role with nothing on the bus is idle; a role session gets the claim rows unchanged (25.773808ms)
✔ agents: working counts what the holder received and has not read; routed alone is not working (25.73112ms)
✔ agents: input needed is the role's open decisions routed to the human, gated first, then oldest (44.674625ms)
✔ agents: working wins over waiting, and Input needed still shows (18.747604ms)
✔ agents: a new holder doesn't inherit what the last one left unread; decisions stay with the role (22.989227ms)
✔ agents: state reads only its own business, even for a run name both use (25.705905ms)
✔ agents: a message the holder released unread is stranded; the next holder gets nothing and cannot read it (26.230589ms)
✔ agents: a message the holder read is not stranded; one the current holder has not read is not listed (22.895582ms)
✔ agents: stranded is oldest first and capped at 50 for a role (105.36242ms)
✔ agents: stranded is per role and per business (43.302129ms)
✔ agents: a message routed to a held role and not yet received is not stranded (18.543389ms)
✔ launch identity is stamped, payload identity is refused and stale holder cannot send (26.161553ms)
✔ decision classes route from policy; gated resolution is human-only, choice and target must match (29.535307ms)
✔ claim exclusion, holder release, gated revoke and rerouting to a new holder are atomic (31.724054ms)
✔ launch events require a human CLI capability; generic emit cannot forge authority events (20.313484ms)
✔ within-role decisions close atomically and invalid options or blocking omissions refuse (21.895459ms)
✔ a recommendation that is not exactly one option key refuses invalid-recommendation after the options (34.326618ms)
✔ observer capabilities read human inbox but cannot mutate or forge launch identity (21.660459ms)
✔ task action subjects and linked decision trail are complete and ordered (31.119535ms)
✔ launch binding is durable and reconnecting requires the identical trusted record (15.187412ms)
✔ business isolation includes inherited object names and cross-business message references (23.257914ms)
✔ authority never transfers between action, run, target, unresolved or replaced role holder (32.848503ms)
✔ task projection uses schema current view, skipping earlier and equal-start polls (17.38437ms)
✔ revocation permanently bars the old run from reclaiming first, including after broker restart (25.895619ms)
✔ empty message references refuse before storage; refusal-evidence failure stays a typed error (19.570396ms)
✔ both arbiters require human resolution when their cross-role route is themselves (37.727859ms)
✔ the shipped coder role raises task.reassign as gated to the human; task.scope.change stays cross-role (29.431582ms)
✔ S1 adapter takes resolved limits and refs, rejects mismatched instance, never mutates input (2.943328ms)
✔ REQ-COS-1, REQ-ROLE-2: the chief of staff ends a run with a reason and no decision, and the reason is logged (30.895851ms)
✔ REQ-ROLE-2: a role without role.end still needs a resolved decision to revoke (35.753284ms)
✔ REQ-ROLE-2: storage accepts a revoke with a reason or a decision and aborts one with neither (23.066663ms)
✔ REQ-ROLE-2: the chief of staff may still cite a decision, and a decision alone is the old path (23.204301ms)
✔ REQ-ROLE-2: a role without role.end still ends a run through a resolved decision, which is recorded and has no reason (29.268033ms)
✔ REQ-ROLE-2: a reason is required, bounded, and never sent with a decision (31.74908ms)
✔ REQ-COS-1: the chief of staff cannot end its own run or the assistant's through role.end (29.356407ms)
✔ REQ-LIMIT-1, REQ-COS-1: organization.set from the chief of staff reaches the shared writer, named as the actor (25.940257ms)
✔ REQ-LIMIT-1: organization.set is refused for every caller whose role file lacks organization.change (32.542532ms)
✔ REQ-LIMIT-1: the chief of staff is logged as allowed, and organization.set takes only the human's keys (39.93075ms)
✔ REQ-ROLE-3, REQ-LIMIT-1: a chief-of-staff change is validated, written and logged like a human's; an invalid one is refused (100.503491ms)
✔ REQ-ROLE-2, REQ-COS-1: the 3c→3d migration lets the chief of staff end a held role with a reason; neither is still refused (78.024708ms)
✔ comment cursors isolate businesses, move only forward and remain append-only (19.630868ms)
✔ only trusted sync writes validated cursors; events and cursor share rollback (16.363071ms)
✔ explicit 3b copy migration preserves evidence and opens as 3d; normal startup never migrates (71.252625ms)
✔ migration refuses active writers and drifted legacy schemas, with no partial upgrade (128.225858ms)
✔ named migration registry chains 3b through 3d to a fixture-only future version in order (75.784739ms)
✔ migration refuses mismatched source metadata without changing schema or evidence (66.555965ms)
✔ offline command migrates only the supplied fixture root and reports usage errors (174.755119ms)
✔ credentials gives human and reader sessions service, account, role, date and state, and nothing else (22.224788ms)
✔ credentials refuses a role session and any argument (19.999647ms)
✔ credentials reads the status on each request, and a broker without one lists none (26.870491ms)
✔ the broker process answers credentials without a token, path or variable in its output, log or store (91.125665ms)
✔ only validated broker references load; returned data and exceptions cannot expose a known token (4.212458ms)
✔ bad file modes, symlinks, repository/data paths, malformed tokens and missing dates refuse (2.386226ms)
✔ expiry refuses use and env references never become client data (0.724428ms)
✔ S1 parsed service refs work, service mismatch refuses, Gitea rotation due is a warning state (1.804819ms)
✔ opaque tokens shorter than 16 characters refuse before use (0.487571ms)
✔ a human resolve sends the raising role one message from the human, and its request is the answer (43.333826ms)
✔ a cross-role decision resolved by its arbiter comes from that role and run, with no request (29.468769ms)
✔ a human withdraw sends one message with the note (26.503631ms)
✔ a human seen before the close leaves the request on the close's answer, not the seen (35.151947ms)
✔ the raising run's withdraw, a supersede, a within-role raise and seen send nothing (36.093875ms)
✔ refused closes send nothing: invalid-choice, decision-closed, not-resolver (32.205194ms)
✔ a raising role nobody holds at the close gets the message on the next claim (28.20074ms)
✔ a human DECISION send keeps its own instruction as the request, not a close's answer (27.053115ms)
✔ roles lists open decisions routed to a role, and never the human inbox (32.386149ms)
✔ recent lists closed decisions of every route, newest close first, with how each closed (34.4117ms)
✔ recent gives the closing event as closed when the decision was seen first (35.689923ms)
✔ recent stops at RECENT rows (114.415214ms)
✔ every row is what trail already gives the same session for that id (45.33563ms)
✔ a role session can't list decisions, so it can't read another role's open decisions (46.032782ms)
✔ decisions is a read: one business, strict arguments, nothing written (53.917363ms)
✔ endLaunch writes session.ended, releases the run claim and kills its capabilities (23.034545ms)
✔ endLaunch refuses an unknown run, a second end and a rebind of the ended run (14.605691ms)
✔ a restarted broker refuses to rebind an ended run; a refused rebind leaves the run unbound (17.436483ms)
✔ endLaunch leaves a claim another run took alone (20.399476ms)
✔ refuse records action.refused against the caller with the code only (15.382884ms)
✔ launches off refuses role.launch with launch-revoked until launches on (20.285954ms)
✔ broker process: launch ops authorize role.launch, record refusals and end runs (118.376196ms)
✔ human proof binds CLI entry, process start and nonce; agents and incomplete ancestry refuse (2.659496ms)
✔ process reader gets own kernel identity without exposing environment values (0.699238ms)
✔ EACCES ancestor environments skip only markers; commands and registered launches still refuse (1.036386ms)
✔ real pid 1 remains inspectable when its environment is protected (0.407212ms)
✔ limits binds actor from human proof and rejects role, reader, tester and forged actor (28.160309ms)
✔ runtime wires verified human settings to the actual shared file writer and actor receipt (53.571078ms)
✔ uncertain organization receipts leave bus startup and reads alone but refuse writes until reconciled (47.930668ms)
✔ within-role sends cite an open gated launch decision without spending it or naming it in grants (30.885692ms)
✔ missing and foreign-business citations refuse and roll back message and grant (21.938091ms)
✔ cross-role sends still need a matching resolved decision and consume it once (28.818866ms)
✔ messages pages all human messages newest first without writes; read state is history (396.283171ms)
✔ messages refuses role callers, invalid paging and cursors outside this human feed (30.247993ms)
✔ views messages passes only its paging arguments through (0.276415ms)
✔ messages isolates businesses even with a valid foreign cursor (20.754868ms)
✔ the assistant sends to the user and to the chief of staff only (38.449145ms)
✔ an assistant send to any other role, or to the tester, is refused and logged (40.432425ms)
✔ only the user and the chief of staff may send to the assistant; the tester is refused by default (41.309501ms)
✔ a sender-side list alone refuses a send to a role that has no list (34.398026ms)
✔ a target-side list alone refuses a send from a role that has no list (36.984568ms)
✔ the sender list is checked before the target list (30.065694ms)
✔ roles without the field behave exactly as before (47.299833ms)
✔ a business with no role definitions at all is unlimited (39.29234ms)
✔ a malformed messaging field refuses at boot (20.574435ms)
✔ the assistant's review.request to a reviewer is refused and logged, even with an approved decision (40.662428ms)
✔ a review.verdict to the assistant from a role outside its receiveFrom is refused and logged (39.617684ms)
✔ review requests and verdicts between roles without the field are unchanged (31.27377ms)
✔ broker process binds trusted launches, offers reader capabilities, refuses human mutation, closes cleanly (92.278443ms)
✔ startup token refusal returns safe code without value or partial listening broker (62.658517ms)
✔ loaded fixture token is absent from socket replies and SQLite, including refusal evidence (85.185144ms)
✔ killed broker leaves an explicit stale lock; another process cannot silently reclaim it (145.434609ms)
✔ trusted host registers later launches; socket clients never have a registration verb (79.175339ms)
✔ runtime excludes declared project roots even when host supplies no repoRoots (45.742883ms)
✔ a refused launch binding leaves the broker and existing capabilities alive; bad protocol stops it (60.638219ms)
✔ a reply is listed for human and reader sessions, and its delivery rows are unchanged (38.673516ms)
✔ the reply still reaches the human message.receive (mosaic talk) after the Console showed it (24.384568ms)
✔ a message with no reply gives an empty list (22.116339ms)
✔ replies come oldest first and stop at REPLIES (93.406423ms)
✔ an id that names no message in the business refuses message-not-found, and the refusal is recorded (34.403351ms)
✔ strict arguments; a role session gets human-required (40.896632ms)
✔ views passes replies through as one call with the id (0.290766ms)
✔ requirements returns the committed list to a holding role session, and no one else (29.757504ms)
✔ the PM's prd.approve with no task, when no id fits, reaches the human's Inbox (19.865047ms)
✔ the PM's non-blocking role.revoke with no task reaches the human's Inbox (24.025224ms)
✔ a push is recorded as the action.allowed for git.push.working on the task, and only the host records one (33.267158ms)
✔ review.request names the latest pushed commit and sends the reviewer a REVIEW-REQUEST (30.495994ms)
✔ review.request says which run holds the reviewer role, or null, and a null request waits for a claim (31.757232ms)
✔ review.request refuses a commit that isn't the latest push, a non-reviewer and a role without the action (43.911281ms)
✔ reviewFetch answers only the addressed reviewer, for the request it cites (31.266201ms)
✔ review.verdict answers one request once, to the role that asked (41.357627ms)
✔ task.close on a pushed task needs an approve verdict for the latest push (48.703263ms)
✔ a refusal the host decided is filed on the task it names, and only a task ref counts (26.228236ms)
✔ v3b prototype refusals, views and append-only mutations (98.92542ms)
✔ gated approval authorizes once, survives store reopen, and fresh approval works (37.321417ms)
✔ another run cannot consume an approval; a failed check leaves it usable (23.660742ms)
✔ two scheduled callers have exactly one grant and one consumed refusal (18.786474ms)
✔ failed commit rolls consumption back; cross-role consumes and within-role stays reusable (40.370916ms)
✔ class drift gated to cross-role refuses before consumption (34.308629ms)
✔ class drift cross-role to gated refuses before consumption (30.782374ms)
✔ class drift gated to within-role refuses before consumption (28.631253ms)
✔ class drift cross-role to within-role refuses before consumption (27.298881ms)
✔ class drift within-role to gated refuses before consumption (24.314312ms)
✔ class drift within-role to cross-role refuses before consumption (21.643172ms)
✔ message.send consumes approval and prevents a later send or authorize (22.047302ms)
✔ role.revoke consumes approval and prevents a later revoke or authorize (21.530776ms)
✔ a business that never switched launches reads allowed with no time and no event (19.55543ms)
✔ launches revoked and then restored read allowed, with the restore time and event id (17.443934ms)
✔ a task closed on the tracker leaves the open list, and total counts the open ones (18.201631ms)
✔ a tombstone is not open (17.661488ms)
✔ the open list stops at the cap and says so; total still counts every open task (22.183501ms)
✔ a role and a reader session read state; arguments, and an unbound cap, refuse (18.641086ms)
✔ a human session reads the same state, with the same cap (24.653856ms)
✔ reading state writes nothing: every table keeps its row count (20.896878ms)
✔ creates private WAL store and excludes a second writer until explicit close (17.133075ms)
✔ rollback is atomic and schema metadata is checked against trusted DDL, not just itself (21.302025ms)
✔ existing empty database and symlink runtime directory refuse, never initialize over damage (22.379434ms)
✔ crash during a transaction recovers no partial event after explicit fixture-only lock removal (78.978841ms)
✔ writer refuses mixed at/read_at forms atomically, even through trusted SQL helpers (14.018098ms)
✔ async transactions refuse before invoking their function (10.278068ms)
✔ the session default fits under the reply cap and pages to the end with next (184.393388ms)
✔ the short form: no description, assignees by role, every listed key (72.431288ms)
✔ each filter, alone and together, and total counts every match (141.508643ms)
✔ task_ref reads one task in full, as the whole board has it (83.95337ms)
✔ no arguments is the read the Console and mosaic make, unchanged (68.191997ms)
✔ an unknown argument or a bad value refuses invalid-request (300.463593ms)
✔ tasks rows carry priority and requirement for every session that reads tasks (52.387232ms)
✔ the read model rebuilds from the ledger: a reopened bus and each trail give the same values (55.105986ms)
✔ an empty business and one with snapshots but no task.created read cleanly (14.147074ms)
✔ recordTask keeps sync reads and a role write apart (28.271461ms)
✔ read_at must be one canonical UTC format, so the projection compares strings safely (12.427738ms)
✔ a bad entry refuses the whole record (12.764877ms)
✔ taskView reads the projection for one business (14.4416ms)
✔ requestTask hands only a holder and a task verb to the handler, and records refusals (25.558773ms)
✔ the server sends task verbs to the adapter with its own timeout; other verbs stay synchronous (274.045632ms)
✔ without an adapter the server refuses every task verb (15.264937ms)
✔ the runtime refuses an invalid adapter and closes a valid one (15.479719ms)
✔ the process loads the S3 adapter from plain-data trackers (110.678642ms)
✔ the tester binds only when the business sets bus.tester to exactly true (82.605543ms)
✔ a business may not name a role tester, so the tester address is never a role (17.872848ms)
✔ every verb outside the tester list refuses tester-refused and is recorded against the tester (33.609694ms)
✔ a tester message is from tester, starts [mos test], and writes no human.input or authority (15.097474ms)
✔ a role answers a tester only in reply to a tester message, and the answer never reaches the human (17.074788ms)
✔ the tester reads the human views a reader reads (21.073675ms)
✔ tester proof binds its own CLI, start time and nonce, refuses a managed run and allows an agent (1.245924ms)
✔ neither CLI's nonce proves the other: each proof is bound to its own CLI and nonce variable (0.373079ms)
✔ testerNonce reads only the tester's variable and gives null on any read error (0.229022ms)
✔ an envelope carries exactly one of cap, human and tester; no tester authenticator refuses (15.673178ms)
✔ tester-cli.mjs binds through the runtime when the switch is on, and refuses off or under a run marker (633.225331ms)
✔ socket capability stamps launch identity; shared views use wire, no SQL client (31.311708ms)
✔ two wire claims serialize; a lost reply never automatically retries (22.691352ms)
✔ malformed, oversized and identity-forging envelopes refuse without echoing input (14.584751ms)
✔ client preserves UTF-8 when a response divides a multibyte character (11.684825ms)
✔ committed mutation followed by dropped reply reports unknown and is never retried (13.54732ms)
ℹ tests 190
ℹ suites 0
ℹ pass 190
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 1373.080014
