chore(deps): bump workspace deps to latest, @agentclientprotocol/sdk to 2.2.0
Merged
OK
17d ago+1027+1027+0123456789012345678901234567890123456789-1065-1065-0123456789012345678901234567890123456789
Every workspace dependency on latest, except the Expo SDK 58 upgrade of apps/mobile-benchmark. agent-device 0.21.35 and playwright-core 1.64.1 were already latest.
@agentclientprotocol/sdk1.7.0 -> 2.2.0 (major). 2.0 removedClientSideConnection/AgentSideConnection, sopackages/acp/src/session.tsandtests/fixtures/scripted-agent.tsnow use theclient()/agent()app API withrequest(methods.agent.session.*). Imports come from@agentclientprotocol/sdk/unstable, because promptusageandusage_updateare unstable in 2.x.- Live ACP adapter pins: claude-agent-acp 0.99.1, codex-acp 2.5.0.
- ai 7.0.159 and all
@ai-sdk/*, oxc 0.156, oxlint 1.90, fallow 3.38,@onkernel/sdk0.139,@limrun/api0.68.1, smolmachines 1.32.4,@types/node26.6.6. - Mobile benchmark: latest Expo SDK 57 patches (expo 57.0.31, react-native 0.86.3), matching
expo install --check. The native fingerprint changes, so CI rebuilds the app. @ai-sdk/openai4.0.106 sends JSON decision state asinput_json(data) instead ofinput_text. Updatedpackages/decision/tests/openai.test.tsto the new wire shape.- fallow 3.34+ treats the
StandardSchemaV1interface and its merged namespace as a duplicate export..fallowrc.jsonignores that one export.
const connection = client({ name: 'e2e' })
.onRequest(methods.client.session.requestPermission, ({ params }) => answerPermission(params, ours, calls, turn.own))
.onNotification(methods.client.session.update, ({ params }) => observe(params, ours, calls, turn, onOwnTool))
.connect(ndJsonStream(stdin, stdout));
await connection.agent.request(methods.agent.session.prompt, { sessionId, prompt });
Left out: Expo SDK 58 (react-native 0.88, react 19.3, Stripe RN 0.84). It's a native upgrade with re-recording, so it gets its own PR.
Verified
Ran it locally: yes
node packages/acp/tests/live/acp.tsagainst real Claude Code (claude-agent-acp 0.99.1, sonnet) and Codex (codex-acp 2.5.0). Both record, then replay. The todo flow, secret login and tool inventory pass. The wrong assertion fails withASSERTION_FAILEDas expected. Usage and cost come through, no secret in artifacts, and none of the agent's own tools ran.pnpm checkgreen.pnpm testgreen apart fromacp > skips actions batched behind the verdict, a flake that is already on main (2/15 full acp runs fail on origin/main).- Not run: OpenAI Decisions API live with
input_json(no key locally), and a native benchmark build (CI builds it per fingerprint).