feat(settings): consolidate AI Models picker into Agent Tools#17
Merged
Conversation
`npm run tauri:dev` was unusable on macOS 14.4+ (and stricter on 15+): the bundled aos-mail-node binary spawned and was immediately SIGKILL'd by `amfid` (Apple Mobile File Integrity), so the sidecar died on launch with `WARN sidecar terminated: signal: Some(9)`. Tauri's shell plugin then logged `sidecar stdin write failed: Broken pipe` and the app sat broken at the splash. Root cause: prepare-node-binary.mjs explicitly strips the Node Foundation signature (`codesign --remove-signature`) with a comment that "Tauri will re-sign with our Developer ID at bundle time." That's true for `tauri build` (via tauri-action's keychain dance), but in `tauri dev` nothing re-signs — the binary stays unsigned, and modern macOS refuses to execute unsigned arm64 Mach-O binaries. Fix: after the strip, add a `codesign --sign - --force` ad-hoc signing step. Ad-hoc signatures satisfy `amfid` for local execution and don't interfere with tauri-action's later `codesign --force` with the real Developer ID cert in production builds — `--force` replaces whatever signature is there. Verified locally: `aos-mail-node --version` now returns v24.14.0 (exit 0) instead of dying with exit 137. The running `tauri dev` session picked up the new binary on its next file-change rebuild and the sidecar booted cleanly: `sidecar: ready, listening on stdin`. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
User asked to move the AI Models section from General to "under the authentication for anthropic in agent tools." Doing so deduplicates a problem that had been growing: General had its own 8-feature model picker, Agent Tools had a 6-feature one (analysis/drafts/refinement/ summary/archiveReady/senderLookup), and changing the same feature in both wrote the same modelConfig key with whichever value got saved last — confusing. Single picker now lives in Agent Tools → AI Models card, directly below the Authentication card that holds the Anthropic API key. Covers all 9 features: analysis, drafts, refinement, calendaring (new in this card), summary, archiveReady, senderLookup, agentDrafter (new), agentChat (new). The General tab's AI Models section is removed. A placeholder comment is left so a future reader doesn't reintroduce it. The Updates card immediately follows, so the General tab now flows straight from Notifications → Updates. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
User request: move the AI Models picker from the General tab to the Agent Tools tab, directly below the Anthropic authentication section.
What was wrong before
Two separate AI Models pickers, both writing to the same `modelConfig` prefs:
Overlap on analysis/drafts/refinement/archiveReady/senderLookup meant whichever tab the user saved last won. Confusing UX, and `calendaring`/`agentDrafter`/`agentChat` could only be configured from one place while `summary` could only be configured from the other.
What this PR does
Single source of truth, lives directly under the Anthropic auth card (Authentication → AI Models). The OpenRouter key input, Free Model loader, and per-feature picker were already grouped together in the Agent Tools AI Models card — this PR just consolidates into that existing structure rather than reorganising it.
Test plan
🤖 Generated with Claude Code