{"version":"2.1.293","anchor":"desktop-3p-config-default-model-effort-always-start-defaul","canonical_anchor":"desktop-3p-config-default-model-effort-always-start-defaul","heading":"New managed settings for Claude Desktop admins: default model, model catalog and more","tier":"use","area":"Desktop App","scope":"org","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293\/e\/desktop-3p-config-default-model-effort-always-start-defaul","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293","markdown":"### New managed settings for Claude Desktop admins: default model, model catalog and more\n\nAdmins of third-party Claude Desktop deployments get managed settings for default model effort, model catalog and Chrome password managers\n\n**What**\n\nThe managed configuration for third-party Claude Desktop deployments has new keys. Managed configuration means settings an administrator sets centrally for everyone in an organisation.\n\n- `defaultModelEffort`\n\n- `alwaysStartWithDefaultModel`\n\n- `modelCatalogEnabled`\n\n- `modelCatalogUrl`\n\n- `claudeInChromePasswordManagersEnabled`, whose value comes from the server\n\nSeveral keys are now marked as applying to the desktop app, and the schema description mentions `x-executionTarget`. The help for `otlpTracesEnabled` now says that, when `otlpEndpoint` is set, this setting alone decides whether traces are exported. Scanning of the credential store also skips its own set of names at the top level of the store.\n\n**Why**\n\nAdministrators of these Desktop deployments get more control over which model people start with and which models they can pick. This is Desktop configuration code that ships inside Claude Code, so it matters mostly to Desktop admins rather than to command-line users.\n\n- Area: Desktop App\n- Tier: Use it now\n- Useful: 4\/5\n- Signal: 3\/5\n- Scope: org\n- Heads-up: no"}