Align avatar video CLI default quality with API

[chasehuh]

Summary

• Change the CLI Avatar Video default quality from standard to plus.
• Update CLI and MCP/tool-registry tests so omitted-quality requests match the current deployed API contract.
• Keep explicit standard, plus, and max quality selection unchanged.

Context

The public API launch contract now defaults Avatar Video generation to plus. The current v0.1.0 CLI release source still defaulted omitted --quality values to standard, which made the install artifact drift from deployed OpenAPI and public docs.

Implementation

• src/lib/quality.ts: updates the single shared DEFAULT_AVATAR_VIDEO_QUALITY constant to plus.
• test/cli.test.ts: updates direct CLI request-body expectations for commands that omit quality.
• test/mcp.test.ts: updates tool execution expectations for omitted quality. Explicit payloads that pass quality: "standard" remain covered and unchanged.

Verification

• pnpm install --frozen-lockfile
• pnpm test -- cli.test.ts passed: 8 files / 119 tests
• pnpm test -- mcp.test.ts passed: 8 files / 119 tests
• pnpm run typecheck passed
• pnpm run build:binary passed
• pnpm run secrets:check passed
• pnpm run test passed: 8 files / 119 tests
• pnpm run build passed
• git diff --check passed

Manual smoke from the built binary:

• ./dist/sume --help keeps MCP/setup out of launched top-level help.
• ./dist/sume avatar-videos create --help shows --quality default plus.
• ./dist/sume --json tools schema avatar-videos.create describes quality as standard|plus|max and defaults to plus.
• ./dist/sume mcp --json, mcp doctor --json, mcp install --agent codex --json, and setup agent --agent codex --json report MCP/agent setup as coming soon without writing config.

Risk / Rollback

Low risk: this changes only the client-side omitted-quality default. Users can still request standard, plus, or max explicitly. Rollback is the one-line constant change if the API default changes again.

Secrets

No secrets or env files are included.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, add credits to your account and enable them for code reviews in your settings.