✔ agents: a held role with nothing on the bus is idle; a role session gets the claim rows unchanged (28.067785ms)
✔ agents: working counts what the holder received and has not read; routed alone is not working (26.453359ms)
✔ agents: input needed is the role's open decisions routed to the human, gated first, then oldest (51.422452ms)
✔ agents: working wins over waiting, and Input needed still shows (19.539458ms)
✔ agents: a new holder doesn't inherit what the last one left unread; decisions stay with the role (23.41594ms)
✔ agents: state reads only its own business, even for a run name both use (24.565002ms)
✔ agents: a message the holder released unread is stranded; the next holder gets nothing and cannot read it (30.168439ms)
✔ agents: a message the holder read is not stranded; one the current holder has not read is not listed (27.767143ms)
✔ agents: stranded is oldest first and capped at 50 for a role (123.006928ms)
✔ agents: stranded is per role and per business (50.523319ms)
✔ agents: a message routed to a held role and not yet received is not stranded (17.219554ms)
✔ launch identity is stamped, payload identity is refused and stale holder cannot send (29.006172ms)
✔ decision classes route from policy; gated resolution is human-only, choice and target must match (36.148326ms)
✔ claim exclusion, holder release, gated revoke and rerouting to a new holder are atomic (36.508171ms)
✔ launch events require a human CLI capability; generic emit cannot forge authority events (20.76218ms)
✔ within-role decisions close atomically and invalid options or blocking omissions refuse (21.514272ms)
✔ a recommendation that is not exactly one option key refuses invalid-recommendation after the options (37.472628ms)
✔ observer capabilities read human inbox but cannot mutate or forge launch identity (24.960147ms)
✔ task action subjects and linked decision trail are complete and ordered (38.870658ms)
✔ launch binding is durable and reconnecting requires the identical trusted record (17.359537ms)
✔ business isolation includes inherited object names and cross-business message references (25.496232ms)
✔ authority never transfers between action, run, target, unresolved or replaced role holder (36.358989ms)
✔ task projection uses schema current view, skipping earlier and equal-start polls (19.996703ms)
✔ revocation permanently bars the old run from reclaiming first, including after broker restart (36.778949ms)
✔ empty message references refuse before storage; refusal-evidence failure stays a typed error (24.360267ms)
✔ both arbiters require human resolution when their cross-role route is themselves (37.685334ms)
✔ the shipped coder role raises task.reassign as gated to the human; task.scope.change stays cross-role (27.954779ms)
✔ S1 adapter takes resolved limits and refs, rejects mismatched instance, never mutates input (2.511939ms)
✔ REQ-COS-1, REQ-ROLE-2: the chief of staff ends a run with a reason and no decision, and the reason is logged (33.752839ms)
✔ REQ-ROLE-2: a role without role.end still needs a resolved decision to revoke (43.254215ms)
✔ REQ-ROLE-2: storage accepts a revoke with a reason or a decision and aborts one with neither (24.2001ms)
✔ REQ-ROLE-2: the chief of staff may still cite a decision, and a decision alone is the old path (24.988236ms)
✔ REQ-ROLE-2: a role without role.end still ends a run through a resolved decision, which is recorded and has no reason (29.642895ms)
✔ REQ-ROLE-2: a reason is required, bounded, and never sent with a decision (32.309147ms)
✔ REQ-COS-1: the chief of staff cannot end its own run or the assistant's through role.end (31.28498ms)
✔ REQ-LIMIT-1, REQ-COS-1: organization.set from the chief of staff reaches the shared writer, named as the actor (28.952573ms)
✔ REQ-LIMIT-1, REQ-COS-1: a refused organization.set logs action.refused and no action.allowed, like a refused limits.set (35.34328ms)
✔ REQ-LIMIT-1: organization.set is refused for every caller whose role file lacks organization.change (35.376971ms)
✔ REQ-LIMIT-1: the chief of staff is logged as allowed, and organization.set takes only the human's keys (46.45152ms)
✔ 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 (92.429545ms)
✔ 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 (81.403732ms)
✔ comment cursors isolate businesses, move only forward and remain append-only (24.98887ms)
✔ only trusted sync writes validated cursors; events and cursor share rollback (15.065434ms)
✔ explicit 3b copy migration preserves evidence and opens as 3d; normal startup never migrates (88.247872ms)
✔ migration refuses active writers and drifted legacy schemas, with no partial upgrade (137.270254ms)
✔ named migration registry chains 3b through 3d to a fixture-only future version in order (91.083961ms)
✔ migration refuses mismatched source metadata without changing schema or evidence (81.833582ms)
✔ offline command migrates only the supplied fixture root and reports usage errors (176.01797ms)
✔ credentials gives human and reader sessions service, account, role, date and state, and nothing else (22.674695ms)
✔ credentials refuses a role session and any argument (21.916526ms)
✔ credentials reads the status on each request, and a broker without one lists none (30.874822ms)
✔ the broker process answers credentials without a token, path or variable in its output, log or store (96.671613ms)
✔ only validated broker references load; returned data and exceptions cannot expose a known token (4.074758ms)
✔ bad file modes, symlinks, repository/data paths, malformed tokens and missing dates refuse (2.876085ms)
✔ expiry refuses use and env references never become client data (0.881236ms)
✔ S1 parsed service refs work, service mismatch refuses, Gitea rotation due is a warning state (1.919445ms)
✔ opaque tokens shorter than 16 characters refuse before use (0.597242ms)
✔ a human resolve sends the raising role one message from the human, and its request is the answer (45.960979ms)
✔ a cross-role decision resolved by its arbiter comes from that role and run, with no request (33.510055ms)
✔ a human withdraw sends one message with the note (29.032256ms)
✔ a human seen before the close leaves the request on the close's answer, not the seen (34.710429ms)
✔ the raising run's withdraw, a supersede, a within-role raise and seen send nothing (40.6562ms)
✔ refused closes send nothing: invalid-choice, decision-closed, not-resolver (40.488037ms)
✔ a raising role nobody holds at the close gets the message on the next claim (34.151803ms)
✔ a human DECISION send keeps its own instruction as the request, not a close's answer (30.72508ms)
✔ roles lists open decisions routed to a role, and never the human inbox (33.142959ms)
✔ recent lists closed decisions of every route, newest close first, with how each closed (44.428015ms)
✔ recent gives the closing event as closed when the decision was seen first (36.081644ms)
✔ recent stops at RECENT rows (123.724583ms)
✔ every row is what trail already gives the same session for that id (51.557192ms)
✔ a role session can't list decisions, so it can't read another role's open decisions (59.726332ms)
✔ decisions is a read: one business, strict arguments, nothing written (58.504551ms)
✔ endLaunch writes session.ended, releases the run claim and kills its capabilities (24.204326ms)
✔ endLaunch refuses an unknown run, a second end and a rebind of the ended run (18.362557ms)
✔ a restarted broker refuses to rebind an ended run; a refused rebind leaves the run unbound (20.494177ms)
✔ endLaunch leaves a claim another run took alone (24.330505ms)
✔ refuse records action.refused against the caller with the code only (15.142932ms)
✔ launches off refuses role.launch with launch-revoked until launches on (21.814041ms)
✔ broker process: launch ops authorize role.launch, record refusals and end runs (130.546923ms)
✔ human proof binds CLI entry, process start and nonce; agents and incomplete ancestry refuse (4.396649ms)
✔ process reader gets own kernel identity without exposing environment values (0.712523ms)
✔ EACCES ancestor environments skip only markers; commands and registered launches still refuse (1.025314ms)
✔ real pid 1 remains inspectable when its environment is protected (0.424489ms)
✔ limits binds actor from human proof and rejects role, reader, tester and forged actor (27.845675ms)
✔ runtime wires verified human settings to the actual shared file writer and actor receipt (61.342522ms)
✔ uncertain organization receipts leave bus startup and reads alone but refuse writes until reconciled (50.156554ms)
✔ within-role sends cite an open gated launch decision without spending it or naming it in grants (33.748572ms)
✔ missing and foreign-business citations refuse and roll back message and grant (26.507855ms)
✔ cross-role sends still need a matching resolved decision and consume it once (33.408388ms)
✔ messages pages all human messages newest first without writes; read state is history (438.908947ms)
✔ messages refuses role callers, invalid paging and cursors outside this human feed (32.330445ms)
✔ views messages passes only its paging arguments through (0.431361ms)
✔ messages isolates businesses even with a valid foreign cursor (22.138374ms)
✔ the assistant sends to the user and to the chief of staff only (44.058172ms)
✔ an assistant send to any other role, or to the tester, is refused and logged (41.006248ms)
✔ only the user and the chief of staff may send to the assistant; the tester is refused by default (44.476419ms)
✔ a sender-side list alone refuses a send to a role that has no list (37.777612ms)
✔ a target-side list alone refuses a send from a role that has no list (41.744617ms)
✔ the sender list is checked before the target list (35.681589ms)
✔ roles without the field behave exactly as before (49.497225ms)
✔ a business with no role definitions at all is unlimited (41.331756ms)
✔ a malformed messaging field refuses at boot (19.159441ms)
✔ the assistant's review.request to a reviewer is refused and logged, even with an approved decision (46.469246ms)
✔ a review.verdict to the assistant from a role outside its receiveFrom is refused and logged (44.478685ms)
✔ review requests and verdicts between roles without the field are unchanged (31.401506ms)
✔ broker process binds trusted launches, offers reader capabilities, refuses human mutation, closes cleanly (98.493646ms)
✔ startup token refusal returns safe code without value or partial listening broker (71.537153ms)
✔ loaded fixture token is absent from socket replies and SQLite, including refusal evidence (97.846946ms)
✔ killed broker leaves an explicit stale lock; another process cannot silently reclaim it (154.006789ms)
✔ trusted host registers later launches; socket clients never have a registration verb (79.906096ms)
✔ runtime excludes declared project roots even when host supplies no repoRoots (49.574724ms)
✔ a refused launch binding leaves the broker and existing capabilities alive; bad protocol stops it (66.774262ms)
✔ a reply is listed for human and reader sessions, and its delivery rows are unchanged (41.706348ms)
✔ the reply still reaches the human message.receive (mosaic talk) after the Console showed it (26.397986ms)
✔ a message with no reply gives an empty list (19.194805ms)
✔ replies come oldest first and stop at REPLIES (104.405804ms)
✔ an id that names no message in the business refuses message-not-found, and the refusal is recorded (38.02522ms)
✔ strict arguments; a role session gets human-required (49.380321ms)
✔ views passes replies through as one call with the id (0.405789ms)
✔ requirements returns the committed list to a holding role session, and no one else (31.12386ms)
✔ the PM's prd.approve with no task, when no id fits, reaches the human's Inbox (21.013698ms)
✔ the PM's non-blocking role.revoke with no task reaches the human's Inbox (23.529477ms)
✔ a push is recorded as the action.allowed for git.push.working on the task, and only the host records one (35.682733ms)
✔ review.request names the latest pushed commit and sends the reviewer a REVIEW-REQUEST (32.725562ms)
✔ review.request says which run holds the reviewer role, or null, and a null request waits for a claim (38.611832ms)
✔ review.request refuses a commit that isn't the latest push, a non-reviewer and a role without the action (48.619672ms)
✔ reviewFetch answers only the addressed reviewer, for the request it cites (36.805365ms)
✔ review.verdict answers one request once, to the role that asked (44.396542ms)
✔ task.close on a pushed task needs an approve verdict for the latest push (54.502798ms)
✔ a refusal the host decided is filed on the task it names, and only a task ref counts (31.233601ms)
✔ v3b prototype refusals, views and append-only mutations (109.262563ms)
✔ gated approval authorizes once, survives store reopen, and fresh approval works (42.584564ms)
✔ another run cannot consume an approval; a failed check leaves it usable (30.377293ms)
✔ two scheduled callers have exactly one grant and one consumed refusal (21.829364ms)
✔ failed commit rolls consumption back; cross-role consumes and within-role stays reusable (49.227372ms)
✔ class drift gated to cross-role refuses before consumption (34.772291ms)
✔ class drift cross-role to gated refuses before consumption (35.102094ms)
✔ class drift gated to within-role refuses before consumption (32.899048ms)
✔ class drift cross-role to within-role refuses before consumption (30.712908ms)
✔ class drift within-role to gated refuses before consumption (22.398805ms)
✔ class drift within-role to cross-role refuses before consumption (20.268256ms)
✔ message.send consumes approval and prevents a later send or authorize (20.263017ms)
✔ role.revoke consumes approval and prevents a later revoke or authorize (20.426937ms)
✔ a business that never switched launches reads allowed with no time and no event (25.323162ms)
✔ launches revoked and then restored read allowed, with the restore time and event id (22.384089ms)
✔ a task closed on the tracker leaves the open list, and total counts the open ones (22.004952ms)
✔ a tombstone is not open (20.590711ms)
✔ the open list stops at the cap and says so; total still counts every open task (26.000304ms)
✔ a role and a reader session read state; arguments, and an unbound cap, refuse (22.012346ms)
✔ a human session reads the same state, with the same cap (24.71103ms)
✔ reading state writes nothing: every table keeps its row count (25.043088ms)
✔ creates private WAL store and excludes a second writer until explicit close (24.768796ms)
✔ rollback is atomic and schema metadata is checked against trusted DDL, not just itself (21.696845ms)
✔ existing empty database and symlink runtime directory refuse, never initialize over damage (25.11103ms)
✔ crash during a transaction recovers no partial event after explicit fixture-only lock removal (71.918076ms)
✔ writer refuses mixed at/read_at forms atomically, even through trusted SQL helpers (16.703623ms)
✔ async transactions refuse before invoking their function (10.606814ms)
✔ the session default fits under the reply cap and pages to the end with next (188.33606ms)
✔ the short form: no description, assignees by role, every listed key (77.774946ms)
✔ each filter, alone and together, and total counts every match (176.438061ms)
✔ task_ref reads one task in full, as the whole board has it (109.474745ms)
✔ no arguments is the read the Console and mosaic make, unchanged (98.424675ms)
✔ an unknown argument or a bad value refuses invalid-request (381.902996ms)
✔ tasks rows carry priority and requirement for every session that reads tasks (55.161926ms)
✔ the read model rebuilds from the ledger: a reopened bus and each trail give the same values (54.942401ms)
✔ an empty business and one with snapshots but no task.created read cleanly (14.549408ms)
✔ recordTask keeps sync reads and a role write apart (29.546254ms)
✔ read_at must be one canonical UTC format, so the projection compares strings safely (14.363921ms)
✔ a bad entry refuses the whole record (12.42493ms)
✔ taskView reads the projection for one business (16.353759ms)
✔ requestTask hands only a holder and a task verb to the handler, and records refusals (26.618468ms)
✔ the server sends task verbs to the adapter with its own timeout; other verbs stay synchronous (275.792315ms)
✔ without an adapter the server refuses every task verb (18.929423ms)
✔ the runtime refuses an invalid adapter and closes a valid one (20.357389ms)
✔ the process loads the S3 adapter from plain-data trackers (136.758006ms)
✔ the tester binds only when the business sets bus.tester to exactly true (82.751545ms)
✔ a business may not name a role tester, so the tester address is never a role (17.137016ms)
✔ every verb outside the tester list refuses tester-refused and is recorded against the tester (34.7707ms)
✔ a tester message is from tester, starts [mos test], and writes no human.input or authority (14.032979ms)
✔ a role answers a tester only in reply to a tester message, and the answer never reaches the human (17.909866ms)
✔ the tester reads the human views a reader reads (20.137198ms)
✔ tester proof binds its own CLI, start time and nonce, refuses a managed run and allows an agent (1.289728ms)
✔ neither CLI's nonce proves the other: each proof is bound to its own CLI and nonce variable (0.405847ms)
✔ testerNonce reads only the tester's variable and gives null on any read error (0.247727ms)
✔ an envelope carries exactly one of cap, human and tester; no tester authenticator refuses (18.864971ms)
✔ tester-cli.mjs binds through the runtime when the switch is on, and refuses off or under a run marker (757.367338ms)
✔ socket capability stamps launch identity; shared views use wire, no SQL client (32.023523ms)
✔ two wire claims serialize; a lost reply never automatically retries (22.292329ms)
✔ malformed, oversized and identity-forging envelopes refuse without echoing input (12.460044ms)
✔ client preserves UTF-8 when a response divides a multibyte character (11.680008ms)
✔ committed mutation followed by dropped reply reports unknown and is never retried (11.684186ms)
ℹ tests 191
ℹ suites 0
ℹ pass 191
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 1559.590174