{"version":"0.2.80","anchor":"new-features","canonical_anchor":"new-features","heading":"New Features","tier":null,"area":null,"scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/0.2.80\/e\/new-features","release_url":"https:\/\/changelogs.core-directive.com\/v\/0.2.80","markdown":"### New Features\n\nOAuth Support for Claude.ai Pro\/Team Users (`CLAUDE_MAX`)\n\n- **New environment variable**: `CLAUDE_MAX` enables OAuth authentication for Claude.ai Pro\/Team users\n- When enabled, adds `user:inference` scope to OAuth permissions, allowing Claude Code to use your Claude.ai account for API access\n- OAuth flow automatically handles token refresh when tokens are near expiration (within 5 minutes)\n- Separate token storage for Claude.ai OAuth vs Console API keys\n\nDeny Rules for Tool Permissions\n\n- **New permission behavior**: In addition to \"allow\" rules, you can now create \"deny\" rules\n- Navigate between Allow\/Deny modes using Tab or arrow keys in the permission rules interface\n- Denied tools will be automatically rejected without prompting\n- Example: Add a deny rule for `Bash` to prevent any shell command execution\n\nManaged Settings Support\n\n- **Enterprise policy control**: System administrators can now configure tool permissions via managed settings\n- Managed settings are loaded from system-level policy files:\n- macOS: `\/Library\/Application Support\/ClaudeCode\/policies.json`\n- Linux: `\/etc\/claude-code\/policies.json`\n- Managed permission rules cannot be modified or deleted by users\n- Rules are displayed with a special indicator showing they're managed by your organization\n\nTool Renaming\n\n- **dispatch_agent** renamed to **AgentTool** for better clarity\n- **ReadNotebook** renamed to **NotebookReadTool** for consistency\n- **LS** renamed to **LSTool** for consistency"}