Zed (AI agent)
Zed Industries · IDE · homepage · zed-industries/zed
1.29.3latest (2026-11-20)
0.229.0minimum safe version
12advisories for Zed (AI agent)
Minimum safe version is set by CVE-2026-44466 (Allowlist Bypass via Bash Arithmetic Expansion in Terminal Tool Permissions, fixed in 0.229.0): the highest fix among Zed (AI agent)'s medium-or-higher advisories. Only advisories for Zed (AI agent) itself are counted.
Advisories
| ID | Title | Severity | Fixed in | Published |
|---|---|---|---|---|
| CVE-2026-44461 | Remote Command Injection via Unquoted Environment Variable Keys (SSH / WSL Remote) in Zed v0.225.9 | High | 0.227.1 | 2026-05-08 |
| CVE-2026-44463 | Allowlist Bypass via Environment Variable Injection in Terminal Tool Permissions | High | 0.229.0 | 2026-05-08 |
| CVE-2026-44462 | Allowlist Bypass via Bash Variable Expansion Chain in Terminal Tool Permissions | Medium | 0.229.0 | 2026-05-08 |
| CVE-2026-44466 | Allowlist Bypass via Bash Arithmetic Expansion in Terminal Tool Permissions · sets minimum safe version | High | 0.229.0 | 2026-05-08 |
| CVE-2026-44465 | Zed IDE Arbitrary Code Execution via untrusted repository with poisoned .git/config | High | 0.227.1 | 2026-05-08 |
| CVE-2026-27976 | Zed Extension Sandbox Escape via Tar Symlink Following | High | 0.224.4 | 2026-02-25 |
| CVE-2026-27967 | Symlink Escape in Agent File Tools | High | 0.225.9 | 2026-02-25 |
| CVE-2026-27800 | Zip Slip Path Traversal in Extension Archive Extraction | High | 0.224.4 | 2026-02-25 |
| CVE-2026-25805 | Parameter Values are not shown for MCP Tool Calls. Users cannot detect tool poisoning. | Medium | 0.219.4 | 2026-02-10 |
| CVE-2025-68433 | Zed IDE MCP Context Server Configuration Arbitrary Code Execution | High | 0.218.2-pre | 2025-12-17 |
Show 2 older advisories
| ID | Title | Severity | Fixed in | Published |
|---|---|---|---|---|
| CVE-2025-68432 | Zed IDE LSP Binary Configuration Arbitrary Code Execution | High | 0.218.2-pre | 2025-12-17 |
| CVE-2025-55012 | AI Agent Remote Code Execution | Critical | 0.197.3 | 2025-08-11 |
Hardening checklist
- Keep auto-update enabled and verify the installed version against the minimum safe version.
- Do not open untrusted repositories with agent mode or auto-approve enabled.
- Restrict MCP servers to an allowlist; review each server's permissions and publisher.
- Never place secrets in project files the agent can read; use a secrets manager or env injection.
Rolling these tools out? See the playbook Secure AI coding tools in your organisation.
Lock it down: AI hardening kit (managed settings and MCP allowlist generators, free, in your browser).
Check your version
Via the public API:
curl -s "https://threatdeck.net/api/v1/tools/zed/check?version=<your-version>" Or ask your agent through the ThreatDeck MCP server (https://threatdeck.net/mcp):
check_ai_tool_version { "slug": "zed", "version": "<your-version>" } Updated 2026-11-22