v0.6.0

What's Changed

• docs(adr): 0059 a worker with real drivers holds a per-user lock (#1189) by @V3RON in #1190
• feat(daemon): a worker with real drivers holds a per-user lock by @V3RON in #1193
• docs(adr): 0060 iOS slim writes launchd overrides while the simulator is shut down by @V3RON in #1194
• feat(ios): slim writes launchd overrides before one boot, reads them back by @V3RON in #1197
• fix(auth): refuse unsafe worker ids at join by @V3RON in #1214
• fix(gateway): drop the proxy auth header and cookie before relaying to a worker by @V3RON in #1213
• fix: owner-only SIMLOCK_HOME and per-request waiting-row ownership by @V3RON in #1217
• fix(cli): escape control characters in remote text and restrict requesterId by @V3RON in #1218
• fix(gateway): a worker cannot crash, stall or corrupt its gateway by @V3RON in #1215
• fix(limits): bound input sizes and make their parsing linear by @V3RON in #1216
• fix(android): passthrough refuses emu avd snapshot except list by @V3RON in #1221
• fix(input): ignore messages after a lease's input is released by @V3RON in #1226
• fix(cli): tunnel drops a connection once both halves ended by @V3RON in #1227
• fix(android): bound the emulator mark read in time and size by @V3RON in #1224
• fix(e2e): capture stand-in logs closed when its stdout fails by @V3RON in #1230
• fix(e2e): a held loopback port never stalls its release by @V3RON in #1231
• fix(stream): a screen stream ends at the identity's expiresAt by @V3RON in #1228
• fix(android): fail the boot when the emulator exits, quoting its error (#1199) by @V3RON in #1232
• feat(mcp): list_devices lists enrolled phones by @leonardoxr in #1120
• chore(agents): manuals reach agents, retired words, fresh checkouts, inline test commands by @V3RON in #1252
• docs(adr): 0061 every virtual device is a copy of a source, and a baseline is a named source by @V3RON in #1201
• feat(ios): every simulator is a copy of a seed, and every virtual device serves one lease by @V3RON in #1210
• feat(android): every emulator is a copy of a seed, booted from its boot snapshot by @V3RON in #1222
• feat(baselines): save, list and remove baselines from held leases by @V3RON in #1235
• feat(leasing): a lease names a baseline and gets a copy of it by @V3RON in #1236
• feat(gateway): route baseline lease requests to a worker that holds the baseline by @V3RON in #1241
• feat(ui): a Baselines page and a device's baseline by @V3RON in #1245
• feat(ios): ios.simulator.audio keeps every simulator silent by @V3RON in #1258
• docs(adr): propose 0063, a big module is one composition root over parts that own their state (#1259) by @V3RON in #1262
• test(hygiene): size test, a code file stays at 1,400 lines and a test file at 2,500 by @V3RON in #1274
• refactor(android): adb-tools.ts, the leaf part for env, commands and device folders by @V3RON in #1276
• refactor(android): ownership.ts, the one check before a stop, load or destroy by @V3RON in #1277
• refactor(android): console-ports.ts owns the console port reservations by @V3RON in #1278
• refactor(android): passthrough-rules.ts holds the adb refusal rules by @V3RON in #1279
• refactor(android): avd-config.ts owns config.ini, the config hash and the marks by @V3RON in #1280
• refactor(android): emulator.ts starts, readies and stops an emulator by @V3RON in #1281
• refactor(android): system-images.ts owns system images and components by @V3RON in #1283
• refactor(android): catalog.ts is a part of the Android driver by @V3RON in #1284
• refactor(android): device-table.ts owns the device states, locks, listManaged and pre-root devices by @V3RON in #1285
• refactor(android): seeds.ts is a part of the Android driver by @V3RON in #1286
• feat(ios): ios.simulator.headless false shows each simulator in a window by @V3RON in #1290
• docs(website): rewrite the user docs in plain, task-first English by @V3RON in #1291
• fix(ui): the lease drawer shows the baseline the lease was made from by @V3RON in #1306
• fix(android): pass -crash-report-mode only to an emulator that lists it by @V3RON in #1304
• fix(daemon): a stopping daemon answers new connections with DAEMON_STOPPING by @V3RON in #1302
• fix(ios): a seed simctl no longer knows counts as deleted (#1296) by @V3RON in #1303
• fix(cli): daemon stop waits until the daemon has exited by @V3RON in #1305
• test: raise five shrink-only line limits in source hygiene by @V3RON in #1307

Full Changelog: v0.5.0...v0.6.0