Subscribe

Hub/AI Security/Claude Code

Claude Code

Anthropic · Official site

7 Yes2 Partial1 No0 Unknown
FeatureStatusNotesSourceChecked
Tool-call approval promptsYesRead-only by default; asks before Bash, file edits and MCP toolsSource2026-11-04
Per-tool allow / denyYesallow / ask / deny rules, incl. mcp__server__tool patternsSource2026-11-04
SandboxingYesOS-level sandbox for Bash: Seatbelt on macOS, bubblewrap on Linux/WSL2Source2026-11-04
OAuth for remote MCPYesOAuth 2.0 for remote HTTP servers; project .mcp.json servers need approvalSource2026-11-04
Verified marketplace listingsNoAnthropic says it cannot verify what third-party plugins contain or that they won't changeSource2026-11-04
Auto-install / auto-update controlPartialPlugins update in the background when a marketplace has auto-update on; toggle per marketplaceSource2026-11-04
Admin allowlistsYesManaged settings: MCP allowlists and denylistsSource2026-11-04
Audit logsPartialOpenTelemetry export of tool decisions and usage; no built-in audit log pageSource2026-11-04
Data retention / training opt-outYesConsumer training choice; commercial terms exclude training; ZDR for eligible orgsSource2026-11-04
Security disclosure pageYesHackerOne program; 36 GHSAs published on anthropics/claude-codeSource2026-11-04

Advisories (39)

  1. Not Ratedresolved
  2. High 7.7patched

    TOCTOU Race in Edit Tool Allows File Writes Outside the Workspace

    Claude Code · CVE-2026-88968 · GHSA-5xjx-9qx6-jcrv

  3. Not Ratedopen
  4. High 7.5patched
  5. High 7.7patched
  6. High 7.7patched
  7. Low 2.0patched
  8. High 8.5patched
  9. Not Ratedresearch
  10. High 7.7patched
  11. Medium 4.4patched
  12. Medium 6.0patched

All 39 on the Claude Code profile →

Across the hub

Full Claude Code profile

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.

You'll finish signing up on beehiiv.