v0.3.0-beta.12

What's Changed

New Features 🎉

• feat(tarko-agent-ui): new layout design (#1553) by @ulivz in 21d59fb
• feat(tarko-agent-server-next): support replacing server global logger (#1554) by @cjraft in 4228662
• feat(tarko-agent-server-next): add initial events for agent when change model (#1552) by @cjraft in 55ae6d7
• feat(tarko-agent): restore events on agent session restore (#1548) by @ulivz in 1282b70
• feat: add @tarko/agent-server-next (#1551) by @cjraft in 043a901
• feat(tarko-agent-ui): disable model selector during agent run (#1520) by @ulivz in 751be28
• feat(tarko-agent-ui): sunsetting plan functionality (#1521) by @ulivz in 5d875e2
• feat(tarko-agent-server): add server.models support (#1488) by @ulivz in 0f4fd41
• feat(model-provider): add automatic claude headers support (#1314) by @ulivz in 160b758
• feat(tarko-mcp-agent): remove deprecated mcpClientVersion support (#1510) by @ulivz in 4f69ba2
• feat(tarko-mcp-agent): adjust defaultConnectionTimeout from 180 to 60 seconds (#1505) by @ulivz in 4db1baf
• feat(tarko): sunsetting model.providers from agent kernel (#1483) by @ulivz in 23002fd
• feat(tarko-agent-ui): support line breaks in LinkReaderRenderer (#1482) by @ulivz in f72e813
• feat(tarko-agent-ui): enhance agent ui config system (#1464) by @cjraft in d2f3633
• feat(tarko-agent-ui): display colored terminal output (#1476) by @ulivz in c9d935b
• feat(agent-tars): emphasize python3 usage in system prompt (#1469) by @ulivz in b457234
• feat(tarko-mcp-agent): defaultConnectionTimeout option (#1462) by @ulivz in f49b2d7

Bug Fixes 🐛

• fix(tarko-agent-ui): support all SSE line separators in streaming (#1568) by @ulivz in bdceb7c
• fix(tarko-agent-ui): tailwind cannot transpile styles from @tarko/ui (#1545) by @ulivz in a1f2c53
• fix(tarko-agent-ui): config issues (#1541) by @cjraft in 34bc5b0
• fix(tarko-agent-ui): navbar overlap on small screens (#1519) by @ulivz in 18e1de6
• fix(model-provider): native providers regression issue (#1503) by @ulivz in a450de0
• fix(tarko-agent): prevent max-iter warning in test by @ulivz in fdbf085
• fix(tarko-agent): should track elapsed time for tool execution errors (#1474) by @ulivz in 6e7dd1f
• fix(tarko-agent): wrong agent_run_end status (#1473) by @ulivz in c3a110b
• fix(tarko-agent-ui): prevent undefined extension in code editor header (#1472) by @ulivz in 9b99a7d
• fix(tarko): preserve logo param in session routing (#1471) by @ulivz in 9bff541

Documentation 📚

• docs(tarko-agent): init documentation (#1502) by @ulivz in b7af1e8
• docs: remove duplicate header (#1559) by @ulivz in 52c0616
• docs: correct typo in mcp documentation (#1516) by @ulivz in 435ea77
• docs: refine useCursor implementation (#1501) by @ulivz in 760072b
• docs: upgrade rspress to v2.0.0-beta.32 (#1495) by @ulivz in 0bac513
• docs(agent-tars): agent api documentation (#1459) by @ulivz in 04cc246
• docs: add @tarko/mcp-agent (#1458) by @ulivz in df34b24

Other Changes

• refactor(gui-agent): refactor gui agent sdk: agent, action-parser and operator (#1550) by @ZhaoHeh in ad242a5
• refactor(tarko-agent-ui): refine ui infrastructure (#1546) by @ulivz in c5f4b6e
• refactor(tarko): sunsetting web socket (#1543) by @ulivz in 22623f9
• refactor(tarko-agent-ui): remove final_answer event handling (#1542) by @ulivz in 4e6b89e
• refactor(tarko-agent-ui): make core low-level components stable (#1540) by @ulivz in 16c8660
• refactor(tarko-agent-ui): remove @headlessui/react (#1539) by @ulivz in bb6f014
• refactor(tarko-agent-ui): migrate stable editor atomics to @tarko/ui (#1537) by @ulivz in e8d7659
• refactor(tarko-agent-ui): remove redundant motion animations (#1530) by @ulivz in 404367f
• refactor(tarko-agent-ui): remove parrot comments (#1528) by @ulivz in 4873fb8
• refactor(tarko-agent-ui): deduplicate muiTheme definitions (#1523) by @ulivz in c603b6e
• refactor(tarko-agent-ui): remove redundant code and simplify components (#1527) by @ulivz in 879e758
• refactor(tarko-agent-ui): unify model display logic (#1525) by @ulivz in 8a50599
• refactor(mcp-agent): improve type safety and docs (#1511) by @ulivz in dfd3ad3
• refactor(agent-tars): improve code architecture and docs (#1498) by @ulivz in 59e4e7b
• refactor(model-provider): improve docs and add tests (#1489) by @ulivz in 0f9a59d
• refactor(tarko): simplify chatpanel state logic (#1381) by @ulivz in b25ab2b
• refactor(tarko): remove language badge from code renderer header (#1470) by @ulivz in 08d3fdc
• chore(tars-stack): release 0.3.0-beta.12 by @ulivz in fb06fcc
• chore: remove unused and outdated files (#1566) by @ulivz in 1336f29
• chore: fix expired lark link (close: #1564) (#1565) by @ulivz in f380287
• chore(agent): clean legacy examples (#1487) by @ulivz in 9c98612
• chore: disable editor.formatOnSave setting (#1486) by @ulivz in 74a5e97
• chore(omni-tars): correct grammar in welcome prompt (#1454) by @ulivz in 4730753
• chore(tars-stack): release 0.3.0-beta.11 (#1453) by @ulivz in eb74340
• ci(pnpm-toolkit): add --auto-create-release-branch support (#1535) by @ulivz in 164c021
• ci(pnpm-toolkit): add canary release support (#1534) by @ulivz in 73b0290
• ci(ptk): extract github username from noreply email (#1456) by @ulivz in e61f855

Full Changelog: v0.3.0-beta.11...v0.3.0-beta.12

v0.3.0-beta.11

What's Changed

From v0.3.0-beta.11, we have mainly introduced the Real-time Thinking Duration Protocol, introduced the MongoDB Provider to Agent Server, supported Tarko Agent UI Builder and CLI, and optimized a lot of Web UI experience.

Here is a simple demo of Thinking Duration:

thinking-duration.mp4

New Features 🎉

• feat(tarko): implement MongoDB provider for agent server (#1450) by @cjraft in b69aa5a
• feat(tarko): agui cli for agent ui builder, see @tarko/agent-ui-cli (#1446) by @ulivz in 7bb9184
• feat(o-agent): update sandbox sdk and gui-agent operator (#1437) by @cjraft in 8e2d7bb
• feat(mcp-servers): support mcp offical registry (#1447) by @ycjcl868 in 5d773cf
• feat(tarko): add navbar logo display options (#1443) by @ulivz in 4b1ed1f
• feat(tarko): init @tarko/agent-ui-builder (#1436) by @ulivz in a99ac0c
• feat(tarko): move workspace navItems from header to navbar (#1441) by @ulivz in 73fa2dc
• feat(tarko): add tabbed file viewer for read_multiple_files tool (#1438) by @ulivz in 88f3568
• feat(gui-agent): improve page visibility detection in AIOBrowser (#1431) by @ZhaoHeh in 230853e
• feat(tarko-agent): thinking duration protocol and modernize thinking ui (#1423) by @ulivz in 094d40e
• feat(tarko): refine collected files (#1422) by @ulivz in 95b1bfb
• feat(tarko): add guiAgent.renderBrowserShell option (#1421) by @ulivz in 5a9d8e4

Bug Fixes 🐛

• fix(tarko): fetch actual remote config instead of local file (#1449) by @ryanroe in 083f842
• fix(tarko): external @tarko/agent-ui-builder in agent-cli build (#1445) by @ulivz in fe579ae
• fix(tarko): improve markdown inline code wrapping (#1439) by @ulivz in df9f553
• fix(tarko): resolve react key spread warning and hooks render issue (#1435) by @ulivz in f3f4bf6
• fix(tarko): make thinking toggle default expanded without initial animation (#1432) by @ulivz in ce0947d
• fix(tarko): prevent frequent api/v1/models calls by memoizing callbacks (#1378) by @ulivz in e07ec41
• fix(tarko): improve scroll-to-bottom indicator edge case handling (#1429) by @ulivz in 50eb9f2
• fix(tarko): prevent duplicate session loading in SessionRouter (#1427) by @ulivz in f96d4ff
• fix(tarko-agent): improve JSON parsing in PromptEngineeringToolCallEngine (close: #1360) (#1361) by @ulivz in b2d5817

Other Changes

• refactor(tarko): rename SessionItemInfo to SessionInfo (#1440) by @ulivz in d1b4d97
• refactor(tarko): rename agent-web-ui to agent-ui (#1434) by @ulivz in 9a4e8f5
• refactor(tarko): simplify code editor components (#1425) by @ulivz in 8b46f6f
• chore(tars-stack): release 0.3.0-beta.11 by @ulivz in be3cfab
• chore(all): fix changelog generation (#1420) by @ulivz in e53360b
• ci: remove tag prefix from release scripts (#1451) by @ulivz in 12ebdba
• ci(ptk): github release (#1428) by @ulivz in cbe3894

Full Changelog: @[email protected]

v0.3.0-beta.10

What's Changed

News ✨

We are thrilled to announce that this version officially introduces support for UI-TARS-2! 🎉

• 🔗 X: https://x.com/TsingYoga/status/1963629621326614940
• 📄 Paper: https://arxiv.org/abs/2509.02544
• 💻 Showcase: https://seed-tars.com/showcase/ui-tars-2

This milestone, together with v0.3.0-beta.9, represents a new beginning for the TARS Stack. 🚀

This release wouldn't have been possible without the dedication and hard work of our incredible Core Contributors:
@cjraft, @ZhaoHeh, @ycjcl868, @skychx, @ulivz, and @helio9cn. 🙌

The journey has just begun—stay tuned for more exciting updates! 💡✨

Demo

omni-tars-demo-v5.mp4

search-osworld-and-generate-a-website.mp4

New Features 🎉

• feat(tarko): limit welcome prompts to 3 with shuffle (#1416) by @ulivz in c6d6791
• feat(tarko): refine all empty state (#1408) by @ulivz in 18dc008
• feat(tarko): add user message auto-scroll in normal mode (#1412) by @ulivz in 2c7f55d
• feat(tarko): enhance slug generation with multilingual support (#1410) by @ulivz in 915c7c5
• feat(tarko): auto-scroll for replay (#1407) by @ulivz in da22a39
• feat(tarko): improve ChatInput UX with conditional help text and home variant (#1406) by @ulivz in 8c38bfc
• feat(tarko): refine thinking animation (#1404) by @ulivz in bae4951
• feat(tarko): refine scroll-to-bottom indicator (#1402) by @ulivz in 3a7d239
• feat(tarko): defaults background to white for html renderer (#1397) by @ulivz in c583e7e

Bug Fixes 🐛

• fix(tarko): prevent auto-scroll on refresh for historical user messages (#1415) by @ulivz in 62df723
• fix(tarko): improve scroll-to-bottom indicator detection (#1411) by @ulivz in 556e3a0
• fix(tarko): improve session UI state management (#1409) by @ulivz in 0391c11
• fix(tarko): scroll-to-bottom indicator session switching and edge cases (#1405) by @ulivz in 442dab8
• fix(tarko): improve markdown link parsing edge cases (#1398) by @ulivz in 24fdf31

Other Changes

• refactor(tarko): remove excessive dots from empty states (#1414) by @ulivz in 074559e
• chore(tars-stack): release 0.3.0-beta.10 by @ulivz in 59b59ef
• chore: fix changelog filter scopes and restore missing entries (#1418) by @ulivz in b3fe00d
• chore(o-agent): update example prompts (#1417) by @ulivz in 3b28c9b
• chore(tarko): enhance code block spacing (#1400) by @ulivz in 1752459
• chore(tars-stack): release 0.3.0-beta.9 (#1396) by @ulivz in da0e22b

Full Changelog: v0.3.0-beta.9...v0.3.0-beta.10

v0.3.0-beta.9

What's Changed

News ✨

We are excited to announce that this version officially introduces support for UI-TARS-2!

• Paper: https://arxiv.org/abs/2509.02544
• Demo: https://seed-tars.com/showcase/ui-tars-2
• X: https://x.com/TsingYoga/status/1963629621326614940

New Features 🎉

• feat(tarko): refine LinkReaderRenderer (#1393) by @ulivz in c985542
• feat(o-agent): temp hack for model thinking (#1395) by @cjraft in 605bf84
• feat(tarko): auto-append replay=1 to share URLs (#1394) by @ulivz in 6a85332
• feat(o-agent): system prompt update (#1392) by @cjraft in b19f9ef
• feat(tarko): disable html rendering in markdown renderer (#1391) by @ulivz in 057a466
• feat(gui-agent): delay 1s before screenshot on aio hybried operator (#1388) by @heh in 79e835a
• feat(o-gui-agent): support ChromeUI gui operation on AIO sandbox (#1383) by @heh in a034369
• feat(tarko): refine behavior of guiAgent.renderGUIAction (#1386) by @ulivz in 94b4c32
• feat(o-agent): update time and proxy instruction in sp (#1384) by @cjraft in 1906ec6
• feat(tarko): add multimodal clipboard paste support (#1379) by @ulivz in 2b40a7c
• feat(tarko): refactor chat panel ui (#1375) by @ulivz in 70c28fa
• feat(tarko): reuse chat input in home page (#1313) by @ulivz in 350364d
• feat(tarko): add model id tooltip to navbar (#1370) by @ulivz in 4da9abb
• feat(o-agent): native think (#1371) by @cjraft in 195c875

Bug Fixes 🐛

• fix(tarko): correct isProcessing state management during agent execution (#1387) by @ulivz in 9d0df70
• fix(tarko): fix markdown link parsing with chinese text (#1358) by @ulivz in 73ca0ca
• fix(tarko): image data missing in workspace (#1373) by @ulivz in 2a79e1d

Other Changes

• refactor(tarko): simplify screenshot display state management (#1390) by @ulivz in d3710ad
• refactor(tarko): remove unnecessary abstraction and redundant state updates (#1380) by @ulivz in dfee2b3
• chore(tars-stack): release 0.3.0-beta.9 by @ulivz in 81a1cfa
• chore(o-agent): disable gui agent screenshot switch and render (#1385) by @ulivz in 2a92348
• chore(ptk): add --no-verify to release commits (#1369) by @ulivz in e19a0f2
• chore(ptk): update release commit scope from agent-tars to tars-stack (#1368) by @ulivz in 25001c7
• chore(all): fix grammar typo (#1367) by @ulivz in 48e40ab
• ci(ptk): handle missing git tags in changelog generation (#1372) by @ulivz in d9eb138

Full Changelog: v0.3.0-beta.8...v0.3.0-beta.9

v0.3.0-beta.8

What's Changed

New Features 🎉

• feat(tarko): implement session state isolation (#1357) by @ulivz in 6f15635
• feat(tarko): unify think rendering with markdown renderer (#1353) by @ulivz in 3a1d53c

Bug Fixes 🐛

• fix(tarko): resolve infinite recursion in layoutModeAtom (#1356) by @ulivz in 91e4016
• fix(tarko): downgrade react-router-dom to v6 for compatibility (#1355) by @ulivz in 5c5887f
• fix(tarko): fallback to beforeActionImage in afterAction strategy to prevent flickering (#1352) by @ulivz in 6190fea
• fix(tarko): hide workspace navigation items in replay mode (#1350) by @ulivz in ccb2262

Other Changes

• chore(tars-stack): release 0.3.0-beta.8 (#1366) by @ulivz in 4f1cd9b
• chore(o-agent): update display texts (#1351) by @ulivz in 8c0f42a

Full Changelog: https://github.com/bytedance/UI-TARS-desktop/compare/@[email protected]...@[email protected]

v0.3.0-beta.7

What's Changed

New Features 🎉

• feat(tarko): remove independent environment input rendering in final state (#1346) by @ulivz in db2515d
• feat(browser-operator): use agent-infra's Hotkey to execute hotkeys (#1343) by @skychx in 0e758f5
• feat(o-gui-agent): temporary solution for getting metadata when screenshot (#1341) by @heh in a56a6c3
• feat(o-agent): enable enableStreamingToolCallEvents (#1340) by @ulivz in 97c937f
• feat(o-gui-agent): support navigate action for new model (#1339) by @heh in 3927337
• feat(tarko): apply RTL only to file-related tools in tool blocks (#1337) by @ulivz in 19bf806
• feat(tarko): trim leading newlines from thinking message content (#1333) by @ulivz in 1e7a553
• feat(omni-gui-agent): adapt tarko's screenshot rendering protocol (#1335) by @heh in cd84f2f
• feat(tarko): only show MessageFooter on final assistant response (#1331) by @ulivz in da3196e
• feat(o-agent): xml parser for agent model (#1330) by @小健 in 80af8c7
• feat(tarko): add math formula rendering support to markdown renderer (#1329) by @ulivz in 1239065
• feat(tarko): show edit_file path in tool call block (#1309) by @ulivz in 28d58d3
• feat(tarko): add url field to screenshot metadata and display in browser shell (#1308) by @ulivz in 4ca0fd9
• feat(tarko): one-click copy raw tool data (#1304) by @ulivz in df001c6
• feat(tarko-web-ui): narrow chat mode (#1298) by @ulivz in f4510f9
• feat(tarko): add gui agent screenshot render strategy config (#1296) by @ulivz in 3730cf6
• feat(agent-tars): strict-typed gui agent procotol (#1295) by @ulivz in 4aa9d78
• feat: enhance streaming for o-agent with improved parsing and processing #1294 (#1294) by @小健 in 4724244
• feat(tarko): switch gui agent to percentage coordinates (#1292) by @ulivz in f56f6fc
• feat(tarko): improve abort button styling (#1290) by @ulivz in 68437e6
• feat(tarko): adjust maxIterations default to 1000 (#1289) by @ulivz in 94e890b
• feat(tarko-web-ui): streaming thinking rendering support (#1284) by @ulivz in ae83d3d
• feat(tarko-agent): add messageId to thinking events for proper session correlation (#1282) by @ulivz in 1fcba4c
• feat(tarko): add codebase metadata to contextual references (#1274) by @ulivz in 6920d83
• feat(tarko): adapt devicePixelRatio from metadata in web ui (#1275) by @ulivz in a728915
• feat(tarko): add metadata field to EnvironmentInputEvent (#1272) by @ulivz in 97ad8aa
• feat(mcp-agent): upgrade mcp-client to 1.2.20 and set 180s timeout (#1271) by @ulivz in 23d73a5
• feat(tarko): support TTFT and TTLT metric (#1232) by @ulivz in bfa2879
• feat(tarko-agent): refine contextual selector (#1134) by @ulivz in aee4bf8
• feat(agent-tars): add static webui config to core (#1266) by @ulivz in 5ba0564

Bug Fixes 🐛

• fix(tarko): persist agent web ui config in share (#1347) by @ulivz in c190d00
• fix(browser): server declares logging capability but doesn't implement method logging/setLevel (#1334) by @charles in 6f537a3
• fix(tarko): browser shell url bar takes full width without spacing (#1327) by @ulivz in 32f71a6
• fix(tarko): unexpected markdown render in generic renderer dark mode (#1324) by @ulivz in 282e306
• fix(tarko): table dark mode styling (#1323) by @ulivz in 173a110
• fix(tarko): move StrategySwitch after ScreenshotDisplay to prevent flicker (#1321) by @ulivz in 91b6053
• fix(tarko): model displayName regression issue (#1315) by @ulivz in 18f34fa
• fix(tarko): replace hardcoded agent name with dynamic config in TerminalOutput (#1306) by @ulivz in f27942e
• fix(tarko): handle open_computer action normalization (#1305) by @ulivz in 871ea58
• fix(tarko): resolve infinite re-render in BrowserControlRenderer hooks (#1303) by @ulivz in 7278561
• fix(tarko): prevent unnecessary environment_input events without contextual references (#1301) by @ulivz in e394343
• fix(agent-server): add safety check for agent.dispose in session cleanup (#1291) by @ulivz in 97ef7ad
• fix(tarko): disable share button during agent execution (#1288) by @ulivz in ba4509b
• fix(tarko-cli): --thinking does not work (#1283) by @ulivz in 03b1d21
• fix(tarko-cli): prevent console interceptor recursion in debug mode (#1279) by @ulivz in 7bcff07
• fix(tarko): improve script execution ui layout and styling (#1268) by @ulivz in fc7a80d
• fix(agent-tars): correct webui property name to webuiConfig (#1267) by @ulivz in 4a5f2fc
• fix(tarko): optimize EditFile title path display (#1246) by @ulivz in 83f8b85

Documentation 📚

• docs(agent-tars): agent hooks (#1277) by @ulivz in 8343182
• docs(agent-tars): preserve tag filter state when navigating back (#1276) by @ulivz in 895c4b3
• docs: fix missing useI18n import in NotFoundLayout (#1265) by @ulivz in 70f67a6
• docs(tarko): add comprehensive event stream documentation (#1242) by @ulivz in 52b44be

Other Changes

• refactor(tarko-web-ui): centralize markdown theme architecture (#1325) by @ulivz in 0067fde
• refactor(common): extract LoadingSpinner and unify modal styles (#1317) by @ulivz in 5c38936
• refactor(tarko): remove meaningless re-exports and restructure web-ui config (#1307) by @ulivz in d35602c
• refactor(tarko-web-ui): extract tooltip props to shared config (#1300) by @ulivz in 9a1b124
• refactor(tarko): remove unused workspace utilities (#1238) by @ulivz in 240595a
• refactor(tarko): extract shared terminal component (#1264) by @ulivz in 37be890
• refactor(tarko): remove over-designed language support (#1263) by @ulivz in 7180405
• refactor(tarko): remove redundant FileRenderer wrapper (#1260) by @ulivz in 0d5a88e
• refactor(tarko): merge EditFileRenderer into DiffRenderer (#1259) by @ulivz in 173a03d
• chore(agent-tars): release 0.3.0-beta.7 (#1348) by @ulivz in 3bdac27
• chore(tarko): remove codeblock action buttons (#1344) by @ulivz in 0f55ce9
• chore(agent): update default layout config (#1311) by @ulivz in 2f4e78d
• chore: only enable route.exclude in production build by @chenhaoli in 50c6923
• chore(tarko): improve gui agent screenshot ui layout and placeholder (#1302) by @ulivz in e083c72
• chore(tarko): replace @ui-tars/operator-browser with local @gui-agent/operator-browser (#1278) by @ulivz in 2c13c04
• chore(mcp-client): release 1.2.20 (#1270) by @ulivz in 5a7200d
• chore(all): unify naming case of webui config (#1269) by @ulivz in 55bb023

Full Changelog: https://github.com/bytedance/UI-TARS-desktop/compare/@[email protected]...@[email protected]

v0.3.0-beta.6

What's Changed

New Features 🎉

• feat(o-agent): add custom timeout for execute_bash tool; remove stop_sequences config (#1256) by @小健 in 5728e0b
• feat(omni-gui-agent): optimize system prompt to use navigate instead of type (#1230) by @heh in c5b4993
• feat(tarko): support top_p configuration for the model (#1247) by @小健 in 9ba651a
• feat(tarko): improve workspace header icons and raw mode spacing by @ulivz in 90a7a8d
• feat(mcp-client): add configurable timeout (#1176) by @ulivz in 858c8c7
• feat(tarko): temporary support for str_replace_editor view command (#1236) by @ulivz in dad2e3d
• feat(tarko): refine str_replace_editor renderer (#1200) by @ulivz in b19de17

Bug Fixes 🐛

• fix(agent-tars): move required deps from devDependencies to dependencies (#1255) by @ulivz in 24e6acf
• fix(tarko): enable line wrapping for command stdout/stderr (#1249) by @ulivz in cda0324
• fix(tarko): update session title in correct metadata structure (#1233) by @ulivz in 94278e5

Documentation 📚

• docs(agent-tars): update video introduction url (#1248) by @ulivz in c81ed80
• docs(tarko-agent): init readme (#1179) by @ulivz in 78fac95

Other Changes

• refactor(tarko): consolidate state atoms (#1237) by @ulivz in 1447009
• chore(agent-tars): release 0.3.0-beta.6 (#1257) by @ulivz in 5569064
• chore(all): standardize the written terminology of Omni-TARS (#1235) by @ulivz in 923785a
• chore(o-tars): using sync mode for execute_bash (#1228) by @ulivz in 06fa5bf
• chore(agent-tars): release 0.3.0-beta.5 (#1227) by @ulivz in 280bbdc

Full Changelog: https://github.com/bytedance/UI-TARS-desktop/compare/@[email protected]...@[email protected]

v0.2.4

What's New

You can also experience the remote versions on Volcano Engine: Computer Operator and Browser Operator.

What's Changed

• feat(ui-tars): sunset UI-TARS-desktop remote operator by @skychx in #1135
• chore(ui-tars): update release version by @ZhaoHeh in #977

Full Changelog: v0.2.3...v0.2.4

v0.3.0-beta.3

What's Changed

New Features 🎉

• feat(tarko): add built-in agents support (#1208) by @ulivz in 2ee2848
• feat(tarko): add webui workspace panels support (#1206) by @ulivz in 04db315
• feat(tarko): add webUIConfig support to AgentConstructor (#1207) by @ulivz in b968bb5
• feat(tarko): add intelligent auto-scroll to chat UI (#1203) by @ulivz in 85b6dd4
• feat(ui-tars): sunset UI-TARS-desktop remote operator (#1135) by @skychx in 21c3910
• feat(tarko): decouple file renderers from GenericResultRenderer (#1201) by @ulivz in 9f586e4
• feat(omni-gui-agent): migrate from local browser to AIO sandbox browser (#1205) by @heh in 3f204bb
• feat(omni-agent): enable gui in omni agent (#1197) by @heh in b564062
• feat(omni-gui-agent): execute screenshot on demand on EachLoopEnd hook (#1195) by @heh in e17643b
• feat(tarko): fully compatible with str_replace_editor (#1189) by @ulivz in 7a4ff74
• feat: upgrade @agent-infra/sandbox package and add health check (#1188) by @小健 in 65b806a
• feat: enhance o-agent with session state management and Jupyter CI support (#1186) by @小健 in fea1084
• feat(tarko): initial support model.displayName (#1163) by @ulivz in 6239834
• feat(tarko): add workspace raw mode display (#1167) by @ulivz in 29826ae
• feat(tarko): add loading states for session creation and switching (#1168) by @ulivz in f551d4c
• feat(tarko): improve JupyterCI tool rendering ui (#1166) by @ulivz in 4d43191
• feat(tarko-cli): load env file baesd on the workspace (#1170) by @小健 in 9482717
• feat(tarko): refine run command semantics (#1158) by @ulivz in 73a79a9
• feat(tarko): add .env file support (#1156) by @小健 in 2279ad9
• feat(mcp-client): add tools and prompts filtering with comprehensive tests (#1155) by @charles in 896274f
• feat(tarko): add agent config viewer (#1153) by @ulivz in 971360b
• feat(agent-tars): support flexible system prompt override (#1151) by @ulivz in d975c30
• feat(tarko): add agent server exclusive mode support (#1149) by @ulivz in acfae7c
• feat(tarko): add workspace config support for instructions.md (#1145) by @ulivz in 1357e48
• feat(agent-cli): auto-detect available port to prevent conflicts (close: #1141) (#1142) by @ulivz in ce9e10b
• feat(mcp): increase default timeout from 10s to 60s (#1139) by @ulivz in 64095e5
• feat(gui-agent): support remote browser operator and update web-ui feature for o-tars gui agent (#1136) by @heh in 2249b98
• feat(o-agent): migrate from omni-tars core to agent-infra sandbox (#1137) by @小健 in cda0a13
• feat(gui-agent): construct operator on demand (#1133) by @heh in b29c1d2
• feat(o-agent): improve configuration and performance optimization (#1131) by @小健 in 61f2b8a
• feat(tarko): o tars adaptation (#1127) by @ulivz in 3ea3053
• feat(tarko): refactor event processor architecture (#1119) by @ulivz in 732aead
• feat(tarko): add raw events state (#1118) by @ulivz in 78c1366
• feat(tarko): display workspace path in workspace header (#1117) by @ulivz in 0b83eee
• feat(navbar): improve width control and model display (#1116) by @ulivz in 4d0b34e
• feat(agent): move aio client to core package, add unit test for parser (#1113) by @小健 in cb7d1f2
• feat(tarko): move model selector from chat input to navbar (#1089) by @ulivz in 28ff271
• feat(tarko): edit_file renderer (#1107) by @ulivz in 855a2da
• feat: enhance code agent and model output adaptation (#1108) by @小健 in d40aa0d
• feat(tarko-agent): add onEachAgentLoopEnd hook (#1111) by @ulivz in 6521137
• feat: add gui agent powered by tarko (#1031) by @heh in c135aa5
• feat(tarko): add LinkReader renderer support (#1099) by @ulivz in 38b9c44
• feat(tarko): optimize time to first token experience (close: #1052) (#1082) by @ulivz in 2faa945
• feat(tarko): support switching model at runtime (close: #1057) (#1058) by @ulivz in 4dcc321
• feat(tarko): add workspace display in navbar (close: #1039) (#1081) by @ulivz in c5a3f9c
• feat(tarko): improve search result relevance scoring (#1079) by @ulivz in 59e8e99
• feat(tarko): optimize navbar space for agent and model display (close: #1076) (#1078) by @ulivz in cb067fd
• feat(tarko-cli): add config logging reminders (close: #1063) by @ulivz in 7fabccf
• feat(omni-tars): migrate gui agent into omni tars (#1071) by @heh in 05bf32b
• feat(gui-agent): add action parser for omni (#1065) by @heh in 153fea9
• feat(omni-tars): refactor AgentPlugin architecture and enhance API integration (#1056) by @小健 in 5805005
• feat(tarko): remove auto scroll behavior from ChatPanel (#1049) by @ulivz in 9887239
• feat(tarko): enhance thinking message ui (#1048) by @ulivz in f40a996
• feat(tarko): mcpServer filter (close: #1045) (#1046) by @ulivz in 77a7fc3
• feat(omni-tars): implement omni-tars multi-agent system (#1047) by @小健 in 1b0c93d
• feat(tarko): tools filter (close: #1041) (#1042) by @ulivz in 1040760
• feat(tarko): experimental contextual selector (#1032) by @ulivz in 478b9a1
• feat(agent-server): handle old workspace schema migration (#1030) by @ulivz in 1057f1b
• feat: seed mcp agent (#1023) by @小健 in 58e599b
• feat(tarko): add @tarko/interface and defineConfig function (#1022) by @ulivz in dc3d2f7
• feat(tarko): agent resolver should respect workspace (#1021) by @ulivz in 52a9fbf
• feat(tarko): webui config and render dynamic ui metadata (#1017) by @ulivz in f794270
• feat(tarko): refine agent module path resolution (#1016) by @ulivz in 03e7a26
• feat(tarko): display agent name in web ui (#1015) by @ulivz in 9cc804c
• feat(tarko): refine package scope (#1013) by @ulivz in 2474789
• feat(tarko): refine workspace resolution (#1011) by @ulivz in 9a7af10
• feat(tarko): refine workspace design (#1008) by @ulivz in 674d67a
• feat(tarko): global directories (#1007) by @ulivz in de40626
• feat(agent-tars): custom agent by @ulivz in 6799ebd
• feat(agent): add dispose api and onDispose hook (#997) by @ulivz in ce2df9e
• feat(agent): add getTools type (#996) by @ulivz in af981e1
• feat(agent-tars-web-ui): simplify replay state (#989) by @ulivz in f865c6d
• feat(agent-tars-server): session read optimization (close: #750) (#974) by @小健 in 68f9805

Bug Fixes 🐛

• fix(tarko): allow workspace panel updates in replay mode (#1202) by @ulivz in 898914f
• fix(tarko): replace hardcoded texts with configurable title (#1174) by @ulivz in 5bd7e26
• fix(tarko): display "Unknown Agent" at initial rendering (#1184) by @ulivz in 6d3b0ca
• fix(tarko): persist agent name in session metadata (#1175) by @ulivz in 436da04
• fix(tarko): handle CLI parameter order for agent argument (#1169) by @ulivz in 2acb378
• fix(tarko): add rollback error handling in sqlite migration (#1147) by @ulivz in 9a49826
• fix(tarko): inline code dark mode text color (#1143) by @ulivz in b37ec25
• fix(tarko): preserve events data during database migration (#1121) by @ulivz in d00fede
• fix(tarko): use plain text rendering for user messages (closes #1103) (#1104) by @ulivz in db45a12
• fix(tarko): improve omni tars search result rendering (close: #1094) (#1096) by @ulivz in ca7bfb5
• fix(tarko): enhance contextual selector with path support and validation (#1077) by @ulivz in fd89a01
• fix(tarko): validate session consistency before panel updates (#1072) by @ulivz in b9cbbd7
• fix(mcp-search): replace node-fetch with native fetch for Node.js 22 (#1069) by @ulivz in e69521d
• fix(omni-tars): add missing super.onAgentLoopEnd() call (#1066) by @ulivz in 9d71442
• fix(agent-tars): directory_tree causes context overflow (close: #969) (#1055) by @ulivz in 9220b25
• fix(agent-tars-cli): sqlite should consider backward compatibility (#1029) by @ulivz in 62f5e05
• fix(tarko): agent cli should pass directories config (#1024) by @ulivz in 0cd72b8
• fix(agent-tars-web-ui): replay does not work (#981) by @ulivz in c39deb9
• fix(mcp-browser): browser mcp screenshot and refactor form_input_fill (#957) by @charles in 26c4131

Documentation 📚

• docs(tarko): enhance agent-server documentation (#1164) by @ulivz in 69f8505
• docs(tarko): improve agent-cli documentation (#1162) by @ulivz in eae3a94
• docs: clarify instructions field behavior (#1059) by @ulivz in 6955142
• docs(agent-tars): fix dead feishu link (close: #1009) (#1010) by @ulivz in 5fdac47
• docs(agent-tars): update showcase tags (#991) by @ulivz in ac40c73
• docs(agent-tars): make showcase public (#988) by @ulivz in aa98838
• docs: update redirects (#983) by @ulivz in ef97e19
• docs: add new redirects (#980) by @ulivz in 6a6b08b
• docs: quick-start.md add links for Volcano Engine's OS Agent (#972) by @skychx in 0e6f62f

Other Changes

• refactor(tarko): flexible condition-based system for tool renderer (#1191) by @ulivz in 00df4a5
• refactor(tarko): some enhancement for gui agent (#1198) by @ulivz in dcf7f7b
• refactor(tarko-web-ui): some enhancements (#1185) by @ulivz in 5d56b38
• refactor(agent-server): refine session item info naming (#1183) by @ulivz in 4a1983b
• refactor(tarko): remote complex mid-layers in workspace renderer (#1120) by @ulivz in 09dcec3
• refactor(tarko): migrate to extensible JSON schema database design (#1122) by @ulivz in 7f6802f
• refactor(tarko): improve agent storage implementation type system (#1025) by @ulivz in 895d8be
• refactor(agent-tars-cli): clean unused dependencies (#1014) by @ulivz in 5244e5d
• refactor(all): refine project structures (#1012) by @ulivz in 74ab1dc
• refactor(agent): sink workspace config to tarko (#998) by @ulivz in a3dca32
• refactor(agent-tars): clean browser control info (#993) by @ulivz in e96f0cd
• refactor(agent-tars-web-ui): comments (#990) by @ulivz in 08369ab
• refactor(mcp-browser): browser_get_markdown (#982) by @charles in cdf385f
• chore(omni-tars): fix dev:agent launch issue by @chenhaoli in b99db7b
• chore(tarko): fix which final environment is shown after non-screensh… (#1209) by @heh in 845cbd0
• chore(gui-agent): fix the missing final screenshot (#1190) by @heh in 0ee8730
• chore(tarko): remove fallback when no screenshot is available by @chenhaoli in 34ad81f
• chore(mcp-browser): ...

v0.2.3

What's New

Bug Fixes 🐛

• Resolved an issue where the browser operator failed to support the HTTP/2 protocol.
• Corrected the default width of the VNC window to ensure proper display.

before

after

Maintenance ⚙️

• Updated the URL for the Volcano Engine OS Agent to point to the new, correct location.

What's Changed

• fix(nut-js): rewrite drag/select by @joey1994 in #909
• fix(browser): remove disable http2 by @ycjcl868 in #936
• feat(ui-tars): change vnc default width and height by @skychx in #955
• docs: quick-start.md add links for Volcano Engine's OS Agent by @skychx in #972

New Contributors

• @joey1994 made their first contribution in #909

Full Changelog: v0.2.2...v0.2.3