AIssistLab

Iteration 01: Release Polish

Goal

Make the current V1 app safer, clearer, and easier to ship without adding new Claude ecosystem surfaces.

Success Criteria

• A new local user can configure workspace paths, select provider mode, rebuild the index, test Claude CLI, and send a first chat from visible UI states.
• Setup Doctor distinguishes blocking errors from quality warnings.
• Chat send behavior is clear when provider or index readiness is not good.
• Docs and diagnostics remain sanitized.

In Scope

• First-run checklist improvements.
• Setup Doctor wording and grouping improvements.
• Chat readiness and empty-state polish.
• Settings copy and status badge polish.
• Browser smoke runbook or checklist.

Out Of Scope

• New provider registry.
• Skill import flows.
• Claude ecosystem inventory.
• LLM-generated diagnosis.

Public Interfaces

No breaking API changes.

Allowed additive fields:

type ReadinessSeverity = "blocking" | "warning" | "optional";

interface DoctorCheck {
  severity?: ReadinessSeverity;
}

interface ChatStatus {
  canSend?: boolean;
  blockingReason?: string | null;
  suggestedAction?: string | null;
}

Likely Modules

• src/lib/settings/doctor.ts
• src/app/api/chat/status/route.ts
• src/app/settings/page.tsx
• src/app/chat/page.tsx
• src/components/layout/Sidebar.tsx
• README.md

Checklist

• Add tests for Doctor severity mapping.
• Add severity to Doctor checks without changing existing status semantics.
• Add tests for chat readiness response when provider is missing, CLI disabled, index missing, index stale, and ready.
• Add optional canSend, blockingReason, and suggestedAction to /api/chat/status.
• Update Chat header/status row to show provider, runtime source, index state, and smoke-test state in compact text.
• Disable or clearly warn before send only when canSend is false.
• Replace generic empty-state prompts with prompts derived from current skill names when available.
• Update Settings first-run checklist to show ordered actions: workspace, index, provider, smoke test, first chat.
• Add sanitized diagnostics copy action or document why it remains export-only.
• Update README with the V1 first-run path.

Tests

• Unit/API:
  • npx --yes tsx src/lib/settings/doctor.test.ts
  • npx --yes tsx src/app/api/chat/status/route.test.ts
• Existing safety:
  • npx --yes tsx src/app/api/settings/local-guards.test.ts
• Final gates:
  • npm run lint
  • npm run build

Browser Smoke

• Open /settings.
• Confirm first-run checklist renders.
• Confirm Setup Doctor top recommendation renders.
• Confirm Test CLI and Open Login buttons keep disabled/loading states.
• Open /chat.
• Confirm readiness row renders.
• Confirm stale/missing index warning offers Rebuild Index.
• Confirm send/retry flow still works with mocked or available provider state.

Privacy Acceptance

• /api/settings/doctor contains no raw home paths, account identifiers, API keys, or raw provider output.
• /api/chat/status contains no raw profile paths or provider secrets.
• Settings UI does not display raw Claude profile folder names.

Checkpoint Recommendation

Checkpoint as feat(v1): polish release readiness only after tests, build, and browser smoke pass.

Implementation Evidence

Implemented in checkpoint 66b56ba feat(release-polish): add chat readiness and setup doctor polish.

Verification completed on 2026-06-14:

• npx --yes tsx src/lib/settings/doctor.test.ts
• npx --yes tsx src/app/api/chat/status/route.test.ts
• npx --yes tsx src/app/api/settings/local-guards.test.ts
• npx --yes tsx src/app/api/index/route.test.ts
• Privacy placeholder scan over README, V1 docs, Settings, Chat, Chat status, and Doctor files.
• npm run lint
• cmd.exe /c npm run build
• Live /settings browser smoke: Doctor rows, severity labels, top actions, first-run checklist, active values, and no recent console errors.
• Live /chat browser smoke: readiness badges, blocked send state, disabled send, suggested prompts, and no recent console errors.
• git diff --check