✔ agents: a held role with nothing on the bus is idle; a role session gets the claim rows unchanged (34.997001ms)
✔ agents: working counts what the holder received and has not read; routed alone is not working (27.816047ms)
✔ agents: input needed is the role's open decisions routed to the human, gated first, then oldest (48.440324ms)
✔ agents: working wins over waiting, and Input needed still shows (26.833379ms)
✔ agents: a new holder doesn't inherit what the last one left unread; decisions stay with the role (24.906339ms)
✔ agents: state reads only its own business, even for a run name both use (24.319609ms)
✔ agents: a message the holder released unread is stranded; the next holder gets nothing and cannot read it (30.639666ms)
✔ agents: a message the holder read is not stranded; one the current holder has not read is not listed (27.872899ms)
✔ agents: stranded is oldest first and capped at 50 for a role (121.86707ms)
✔ agents: stranded is per role and per business (49.736471ms)
✔ agents: a message routed to a held role and not yet received is not stranded (19.598304ms)
✔ launch identity is stamped, payload identity is refused and stale holder cannot send (31.669349ms)
✔ decision classes route from policy; gated resolution is human-only, choice and target must match (35.829036ms)
✔ claim exclusion, holder release, gated revoke and rerouting to a new holder are atomic (35.519266ms)
✔ launch events require a human CLI capability; generic emit cannot forge authority events (22.266483ms)
✔ within-role decisions close atomically and invalid options or blocking omissions refuse (25.84336ms)
✔ a recommendation that is not exactly one option key refuses invalid-recommendation after the options (37.071603ms)
✔ observer capabilities read human inbox but cannot mutate or forge launch identity (25.051182ms)
✔ task action subjects and linked decision trail are complete and ordered (37.949803ms)
✔ launch binding is durable and reconnecting requires the identical trusted record (17.860968ms)
✔ business isolation includes inherited object names and cross-business message references (25.500445ms)
✔ authority never transfers between action, run, target, unresolved or replaced role holder (37.723566ms)
✔ task projection uses schema current view, skipping earlier and equal-start polls (21.211621ms)
✔ revocation permanently bars the old run from reclaiming first, including after broker restart (34.379018ms)
✔ empty message references refuse before storage; refusal-evidence failure stays a typed error (25.530104ms)
✔ both arbiters require human resolution when their cross-role route is themselves (37.406048ms)
✔ the shipped coder role raises task.reassign as gated to the human; task.scope.change stays cross-role (48.457597ms)
✔ S1 adapter takes resolved limits and refs, rejects mismatched instance, never mutates input (2.379809ms)
✔ REQ-COS-1, REQ-ROLE-2: the chief of staff ends a run with a reason and no decision, and the reason is logged (38.677993ms)
✔ REQ-ROLE-2: a role without role.end still needs a resolved decision to revoke (37.317041ms)
✔ REQ-ROLE-2: storage accepts a revoke with a reason or a decision and aborts one with neither (25.932484ms)
✔ REQ-ROLE-2: the chief of staff may still cite a decision, and a decision alone is the old path (30.299568ms)
✔ REQ-ROLE-2: a role without role.end still ends a run through a resolved decision, which is recorded and has no reason (28.63359ms)
✔ REQ-ROLE-2: a reason is required, bounded, and never sent with a decision (33.228942ms)
✔ REQ-COS-1: the chief of staff cannot end its own run or the assistant's through role.end (30.945834ms)
✔ REQ-LIMIT-1, REQ-COS-1: organization.set from the chief of staff reaches the shared writer, named as the actor (26.873118ms)
✔ REQ-LIMIT-1, REQ-COS-1: a refused organization.set logs action.refused and no action.allowed, like a refused limits.set (35.258889ms)
✔ REQ-LIMIT-1: organization.set is refused for every caller whose role file lacks organization.change (35.832096ms)
✔ REQ-LIMIT-1: the chief of staff is logged as allowed, and organization.set takes only the human's keys (47.41763ms)
✔ 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 (121.610795ms)
✔ 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 (110.467639ms)
✔ comment cursors isolate businesses, move only forward and remain append-only (27.018165ms)
✔ only trusted sync writes validated cursors; events and cursor share rollback (17.643369ms)
✔ explicit 3b copy migration preserves evidence and opens as 3d; normal startup never migrates (87.946186ms)
✔ migration refuses active writers and drifted legacy schemas, with no partial upgrade (142.441372ms)
✔ named migration registry chains 3b through 3d to a fixture-only future version in order (87.535444ms)
✔ migration refuses mismatched source metadata without changing schema or evidence (73.526208ms)
✔ offline command migrates only the supplied fixture root and reports usage errors (257.553608ms)
✔ credentials gives human and reader sessions service, account, role, date and state, and nothing else (22.40498ms)
✔ credentials refuses a role session and any argument (24.61ms)
✔ credentials reads the status on each request, and a broker without one lists none (26.213354ms)
✔ the broker process answers credentials without a token, path or variable in its output, log or store (95.221185ms)
✔ only validated broker references load; returned data and exceptions cannot expose a known token (4.742272ms)
✔ bad file modes, symlinks, repository/data paths, malformed tokens and missing dates refuse (3.110598ms)
✔ expiry refuses use and env references never become client data (1.161965ms)
✔ S1 parsed service refs work, service mismatch refuses, Gitea rotation due is a warning state (2.086156ms)
✔ opaque tokens shorter than 16 characters refuse before use (0.644508ms)
✔ a human resolve sends the raising role one message from the human, and its request is the answer (53.246916ms)
✔ a cross-role decision resolved by its arbiter comes from that role and run, with no request (30.158881ms)
✔ a human withdraw sends one message with the note (28.852652ms)
✔ a human seen before the close leaves the request on the close's answer, not the seen (40.420795ms)
✔ the raising run's withdraw, a supersede, a within-role raise and seen send nothing (36.276493ms)
✔ refused closes send nothing: invalid-choice, decision-closed, not-resolver (38.695525ms)
✔ a raising role nobody holds at the close gets the message on the next claim (33.628259ms)
✔ a human DECISION send keeps its own instruction as the request, not a close's answer (30.667439ms)
✔ roles lists open decisions routed to a role, and never the human inbox (38.483739ms)
✔ recent lists closed decisions of every route, newest close first, with how each closed (38.817244ms)
✔ recent gives the closing event as closed when the decision was seen first (43.039187ms)
✔ recent stops at RECENT rows (130.044698ms)
✔ every row is what trail already gives the same session for that id (49.769987ms)
✔ a role session can't list decisions, so it can't read another role's open decisions (55.724765ms)
✔ decisions is a read: one business, strict arguments, nothing written (62.64222ms)
✔ endLaunch writes session.ended, releases the run claim and kills its capabilities (27.013788ms)
✔ endLaunch refuses an unknown run, a second end and a rebind of the ended run (16.783697ms)
✔ a restarted broker refuses to rebind an ended run; a refused rebind leaves the run unbound (20.954199ms)
✔ endLaunch leaves a claim another run took alone (20.919714ms)
✔ refuse records action.refused against the caller with the code only (17.280612ms)
✔ launches off refuses role.launch with launch-revoked until launches on (28.814771ms)
✔ broker process: launch ops authorize role.launch, record refusals and end runs (120.263602ms)
✔ human proof binds CLI entry, process start and nonce; agents and incomplete ancestry refuse (2.468631ms)
✔ process reader gets own kernel identity without exposing environment values (0.732673ms)
✔ EACCES ancestor environments skip only markers; commands and registered launches still refuse (1.004774ms)
✔ real pid 1 remains inspectable when its environment is protected (0.412298ms)
✔ limits binds actor from human proof and rejects role, reader, tester and forged actor (31.59597ms)
✔ runtime wires verified human settings to the actual shared file writer and actor receipt (57.868437ms)
✔ uncertain organization receipts leave bus startup and reads alone but refuse writes until reconciled (51.922006ms)
✔ within-role sends cite an open gated launch decision without spending it or naming it in grants (35.127929ms)
✔ missing and foreign-business citations refuse and roll back message and grant (24.486821ms)
✔ cross-role sends still need a matching resolved decision and consume it once (34.477943ms)
✔ messages pages all human messages newest first without writes; read state is history (480.397952ms)
✔ messages refuses role callers, invalid paging and cursors outside this human feed (49.025963ms)
✔ views messages passes only its paging arguments through (0.414697ms)
✔ messages isolates businesses even with a valid foreign cursor (23.567944ms)
✔ the assistant sends to the user and to the chief of staff only (43.536628ms)
✔ an assistant send to any other role, or to the tester, is refused and logged (46.314328ms)
✔ only the user and the chief of staff may send to the assistant; the tester is refused by default (41.687488ms)
✔ a sender-side list alone refuses a send to a role that has no list (38.738742ms)
✔ a target-side list alone refuses a send from a role that has no list (39.943353ms)
✔ the sender list is checked before the target list (35.437386ms)
✔ roles without the field behave exactly as before (55.008816ms)
✔ a business with no role definitions at all is unlimited (44.226068ms)
✔ a malformed messaging field refuses at boot (20.368571ms)
✔ the assistant's review.request to a reviewer is refused and logged, even with an approved decision (67.363763ms)
✔ a review.verdict to the assistant from a role outside its receiveFrom is refused and logged (57.637528ms)
✔ review requests and verdicts between roles without the field are unchanged (40.64322ms)
✔ broker process binds trusted launches, offers reader capabilities, refuses human mutation, closes cleanly (102.584914ms)
✔ startup token refusal returns safe code without value or partial listening broker (67.364951ms)
✔ loaded fixture token is absent from socket replies and SQLite, including refusal evidence (97.324909ms)
✔ killed broker leaves an explicit stale lock; another process cannot silently reclaim it (183.279479ms)
✔ trusted host registers later launches; socket clients never have a registration verb (104.033723ms)
✔ runtime excludes declared project roots even when host supplies no repoRoots (66.038657ms)
✔ a refused launch binding leaves the broker and existing capabilities alive; bad protocol stops it (78.770164ms)
✔ a reply is listed for human and reader sessions, and its delivery rows are unchanged (52.813417ms)
✔ the reply still reaches the human message.receive (mosaic talk) after the Console showed it (26.69298ms)
✔ a message with no reply gives an empty list (22.371248ms)
✔ replies come oldest first and stop at REPLIES (100.926024ms)
✔ an id that names no message in the business refuses message-not-found, and the refusal is recorded (38.517977ms)
✔ strict arguments; a role session gets human-required (47.710319ms)
✔ views passes replies through as one call with the id (0.308486ms)
✔ requirements returns the committed list to a holding role session, and no one else (38.963837ms)
✔ the PM's prd.approve with no task, when no id fits, reaches the human's Inbox (22.190993ms)
✔ the PM's non-blocking role.revoke with no task reaches the human's Inbox (23.750162ms)
✔ a push is recorded as the action.allowed for git.push.working on the task, and only the host records one (38.859239ms)
✔ review.request names the latest pushed commit and sends the reviewer a REVIEW-REQUEST (30.470161ms)
✔ review.request says which run holds the reviewer role, or null, and a null request waits for a claim (39.335324ms)
✔ review.request refuses a commit that isn't the latest push, a non-reviewer and a role without the action (49.6511ms)
✔ reviewFetch answers only the addressed reviewer, for the request it cites (32.094238ms)
✔ review.verdict answers one request once, to the role that asked (42.423421ms)
✔ task.close on a pushed task needs an approve verdict for the latest push (86.095964ms)
✔ a refusal the host decided is filed on the task it names, and only a task ref counts (39.91629ms)
✔ v3b prototype refusals, views and append-only mutations (108.635596ms)
✔ gated approval authorizes once, survives store reopen, and fresh approval works (38.829608ms)
✔ another run cannot consume an approval; a failed check leaves it usable (29.229112ms)
✔ two scheduled callers have exactly one grant and one consumed refusal (22.006753ms)
✔ failed commit rolls consumption back; cross-role consumes and within-role stays reusable (50.760827ms)
✔ class drift gated to cross-role refuses before consumption (37.044283ms)
✔ class drift cross-role to gated refuses before consumption (46.044143ms)
✔ class drift gated to within-role refuses before consumption (50.128298ms)
✔ class drift cross-role to within-role refuses before consumption (40.383887ms)
✔ class drift within-role to gated refuses before consumption (30.427344ms)
✔ class drift within-role to cross-role refuses before consumption (25.260499ms)
✔ message.send consumes approval and prevents a later send or authorize (24.802735ms)
✔ role.revoke consumes approval and prevents a later revoke or authorize (27.959887ms)
✔ a business that never switched launches reads allowed with no time and no event (23.691368ms)
✔ launches revoked and then restored read allowed, with the restore time and event id (22.218843ms)
✔ a task closed on the tracker leaves the open list, and total counts the open ones (20.10714ms)
✔ a tombstone is not open (17.340197ms)
✔ the open list stops at the cap and says so; total still counts every open task (25.727725ms)
✔ a role and a reader session read state; arguments, and an unbound cap, refuse (21.757222ms)
✔ a human session reads the same state, with the same cap (27.257155ms)
✔ reading state writes nothing: every table keeps its row count (38.808126ms)
✔ creates private WAL store and excludes a second writer until explicit close (19.94566ms)
✔ rollback is atomic and schema metadata is checked against trusted DDL, not just itself (24.589156ms)
✔ existing empty database and symlink runtime directory refuse, never initialize over damage (28.657033ms)
✔ crash during a transaction recovers no partial event after explicit fixture-only lock removal (97.475957ms)
✔ writer refuses mixed at/read_at forms atomically, even through trusted SQL helpers (38.2227ms)
✔ async transactions refuse before invoking their function (18.236254ms)
✔ the session default fits under the reply cap and pages to the end with next (255.025123ms)
✔ the short form: no description, assignees by role, every listed key (106.506372ms)
✔ each filter, alone and together, and total counts every match (200.111922ms)
✔ task_ref reads one task in full, as the whole board has it (122.826393ms)
✔ no arguments is the read the Console and mosaic make, unchanged (106.27715ms)
✔ an unknown argument or a bad value refuses invalid-request (376.655384ms)
✔ tasks rows carry priority and requirement for every session that reads tasks (61.67629ms)
✔ the read model rebuilds from the ledger: a reopened bus and each trail give the same values (81.86796ms)
✔ an empty business and one with snapshots but no task.created read cleanly (19.211114ms)
✔ recordTask keeps sync reads and a role write apart (48.763246ms)
✔ read_at must be one canonical UTC format, so the projection compares strings safely (18.035046ms)
✔ a bad entry refuses the whole record (21.230151ms)
✔ taskView reads the projection for one business (18.995745ms)
✔ requestTask hands only a holder and a task verb to the handler, and records refusals (32.645469ms)
✔ the server sends task verbs to the adapter with its own timeout; other verbs stay synchronous (279.726403ms)
✔ without an adapter the server refuses every task verb (20.351847ms)
✔ the runtime refuses an invalid adapter and closes a valid one (23.810309ms)
✔ the process loads the S3 adapter from plain-data trackers (158.968104ms)
✔ the tester binds only when the business sets bus.tester to exactly true (108.321243ms)
✔ a business may not name a role tester, so the tester address is never a role (22.811944ms)
✔ every verb outside the tester list refuses tester-refused and is recorded against the tester (39.859438ms)
✔ a tester message is from tester, starts [mos test], and writes no human.input or authority (16.297992ms)
✔ a role answers a tester only in reply to a tester message, and the answer never reaches the human (21.53131ms)
✔ the tester reads the human views a reader reads (27.956568ms)
✔ tester proof binds its own CLI, start time and nonce, refuses a managed run and allows an agent (2.167542ms)
✔ neither CLI's nonce proves the other: each proof is bound to its own CLI and nonce variable (0.738694ms)
✔ testerNonce reads only the tester's variable and gives null on any read error (0.251703ms)
✔ an envelope carries exactly one of cap, human and tester; no tester authenticator refuses (18.826693ms)
✔ tester-cli.mjs binds through the runtime when the switch is on, and refuses off or under a run marker (800.99161ms)
✔ socket capability stamps launch identity; shared views use wire, no SQL client (43.416379ms)
✔ two wire claims serialize; a lost reply never automatically retries (27.602582ms)
✔ malformed, oversized and identity-forging envelopes refuse without echoing input (16.193787ms)
✔ client preserves UTF-8 when a response divides a multibyte character (11.981596ms)
✔ committed mutation followed by dropped reply reports unknown and is never retried (17.947423ms)
ℹ tests 191
ℹ suites 0
ℹ pass 191
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 1699.136397
