Remove low-value rendering, dependency-owned and script tests together with orphaned test scaffolding and obsolete wrappers. Keep state-driven visual coverage and release screenshot workflows.
Preserve incoming message origins on preparation failures, refresh tool presentation fixtures, and move shared SSH import coverage out of the iOS host test bundle to avoid duplicate static types.
Relocate the host icon input manifest, remove the heuristic AppStrings checker, and simplify localization CI.
Integrate AgentKit 2.0.1 with shared scratch storage, whole-line tool policies, generated capture paths, and external MCP projection bypass. Update ssh-exec instructions and document the complete tool policy inventory.
Keep compact expansion controls left aligned and display the AgentKit truncation marker above them on both platforms. Cover narrow and wide native layouts, both button states, and marker ordering.
Split command execution into run, output, stdin, cancel and list tools for the assistant and MCP. Preserve bounded output tails and use conversation-scoped handles and secret storage with AgentKit 1.20.0.
Add the built-in ssh-exec skill with timestamped log redirection and text/secret stdin workflows. Expand long tool output without adjacent copy controls, update localized presentation, and refresh regression coverage and the website tool name.
BREAKING CHANGE: Replace ssh_exec and ssh_wait with ssh_exec_run and ssh_exec_output. Move stdin to ssh_exec_stdin and remove execution timeout and cwd arguments. Use run_in_background to control foreground execution.
Integrate AgentKit 0.18.1 with native authorization on macOS and iOS, device-local OAuth credentials, V43 configuration persistence, and Assistant authorization prompts.
Add HTTPS callback metadata, keep native callbacks safe across actor boundaries, and distinguish authorization setup failures from cancellation. Extend the external MCP server to negotiate 2025-03-26.
Add host-owned chat creation, model discovery, messaging, reading, and waiting with durable delivery receipts. Keep tool presentation compact, mark agent-authored messages, and prevent edits that would remove tool activity.
Adopt the signed AgentKit 0.16.0 release in the package requirement, resolved revision, and submodule.
Review every 10 turns or 15 non-memory tool calls and save explicit corrections promptly. Invalidate outdated learning after lifecycle and history changes, and prevent older correction retries from overwriting newer corrections.
Replace capacity bars with entry and character counts, retain per-entry and model-input budgets, and migrate away from frozen snapshots. Upgrade the AgentKit package and submodule to signed release 0.15.0.
Allow optional API keys in desktop and mobile settings and share credential
resolution across model requests. Missing stored credentials remain errors.
Enable user-configured HTTP endpoints and pin signed AgentKit 0.13.1, which
omits empty authentication headers for probes and inference.
Validation: AgentKit's 216 tests and iOS's 461 tests passed, along with Swift
formatting, localization checks, native screenshots, and real HTTP smoke tests
on both apps. macOS passed 1419 of 1422 tests initially; existing Metrics timing,
waveform PNG comparison, and MCP upload runner failures were observed. MCP and
Metrics passed a focused rerun, while Metrics remains intermittent and waveform
comparison still fails. iOS simulator LAN-IP access timed out; desktop LAN-IP
and both platforms' localhost, loopback-IP, and local-domain requests passed.
Keep previously installed starter skills as ordinary user-owned records. Provide hidden read-only Skill Creator and Skill Installer instructions, use AgentKit 0.13.0 for reviewed complete GitHub package installation, and remove skill seeding and Reset.
Move Hermes Memory below MCP, use Enabled for its switch, and localize its Chinese title as persistent memory. Preserve existing user edits and synchronize user skill packages through the existing atomic store.
Validation: AgentKit 213 tests and iOS 452 tests passed. macOS app 1183 tests passed; shared frameworks passed 231/232 with the existing SpeechWaveformViewTests PNG byte-equality failure. Related skill, sync, localization and UI checks passed. Strict Swift formatting, Chinese localization checks, and git diff --check passed.
Add opt-in GRDB memory with frozen conversation snapshots, background learning, session search, and a user-only Destroy action on both platforms. Persist and synchronize complete skill packages with the new AgentKit management tools.
Pin signed AgentKit 0.12.3 and GRDB 7.11.1. Add the V18 package migration, regression coverage, bilingual UI snapshots, and architecture records.
Validation: AgentKit and iOS suites, focused macOS skills/memory/architecture/localization tests, Swift formatting, and Chinese localization checks passed. Full macOS runs exposed intermittent existing waveform-rendering and metrics-timing failures.
Kaede Akino