Hub/AI Security/Claude apps (claude.ai / Desktop)
Claude apps (claude.ai / Desktop)
Anthropic · Official site
4 Yes0 Partial0 No6 Unknown
| Feature | Status | Notes | Source | Checked |
|---|---|---|---|---|
| Tool-call approval prompts | Unknown | Not found in official docs checked 2026-11-04 | — | 2026-11-04 |
| Per-tool allow / deny | Unknown | Not found in official docs checked 2026-11-04 | — | 2026-11-04 |
| Sandboxing | Unknown | Not found in official docs checked 2026-11-04 | — | 2026-11-04 |
| OAuth for remote MCP | Yes | Custom connectors: recommended client, automatic registration or your own OAuth client | Source | 2026-11-04 |
| Verified marketplace listings | Unknown | Not found in official docs checked 2026-11-04 | — | 2026-11-04 |
| Auto-install / auto-update control | Unknown | Not found in official docs checked 2026-11-04 | — | 2026-11-04 |
| Admin allowlists | Yes | On Team/Enterprise only Owners can add custom connectors | Source | 2026-11-04 |
| Audit logs | Yes | Audit log export for Team and Enterprise | Source | 2026-11-04 |
| Data retention / training opt-out | Yes | Consumer accounts choose whether chats train models; commercial data excluded | Source | 2026-11-04 |
| Security disclosure page | Unknown | Not found in official docs checked 2026-11-04 | — | 2026-11-04 |
Advisories (1)
Across the hub
Weekly, free
What changed in AI integrations
One short email when MCP clients, plugin ecosystems, IDE integrations, AI plan prices or security advisories change. Every item is sourced.