Figma MCP version 0.6.3 remediation guidance
Advisory/Mitigation
Summary
Hide ▲
Show ▼
Users of Figma MCP were told to upgrade to version 0.6.3 or higher to reduce exposure to a command-injection flaw that could enable remote code execution. The remediation also calls for an audit of vulnerable systems so operators can find affected deployments. Teams are further advised to review logs for suspicious command execution patterns to spot possible abuse.
Related Happenings
MCP STDIO arbitrary command execution security flaw
Vulnerability
First: 16.04.2026 12:40
Last: 16.04.2026 12:40
Sources 1
About this happening:
A **critical MCP flaw** in the **STDIO interface** can trigger **arbitrary command execution**, putting **connected AI systems** at risk of **data exposure** and **system takeover...
MCP STDIO arbitrary command execution security flaw
VulnerabilityAbout this happening: A **critical MCP flaw** in the **STDIO interface** can trigger **arbitrary command execution**, putting **connected AI systems** at risk of **data exposure** and **system takeover...
Anthropic security patch release for CVE-2025-68143
Security Patch Release
First: 20.01.2026 17:01
Last: 20.01.2026 17:01
Sources 1
About this happening:
**Anthropic** released **December 2025** fixes for **mcp-server-git** vulnerabilities affecting default installations, closing flaws that could enable **prompt injection**, **code...
Anthropic security patch release for CVE-2025-68143
Security Patch ReleaseAbout this happening: **Anthropic** released **December 2025** fixes for **mcp-server-git** vulnerabilities affecting default installations, closing flaws that could enable **prompt injection**, **code...
Mcp-server-git path traversal and argument injection flaws (multiple vulnerabilities)
Vulnerability
First: 20.01.2026 15:55
Last: 20.01.2026 15:55
Sources 1
About this happening:
**Anthropic's mcp-server-git** now has **three disclosed vulnerabilities** that can enable **arbitrary file read/delete** and **code execution** in affected deployments. The issue...
Mcp-server-git path traversal and argument injection flaws (multiple vulnerabilities)
VulnerabilityAbout this happening: **Anthropic's mcp-server-git** now has **three disclosed vulnerabilities** that can enable **arbitrary file read/delete** and **code execution** in affected deployments. The issue...
Figma-developer-mcp MCP server command injection (CVE-2025-53967)
Vulnerability
First: 08.10.2025 13:58
Last: 08.10.2025 13:58
Sources 1
How related:
the "figma-developer-mcp" npm package has a bug (CVE-2025-53967; CVSS 7.5) that allows attackers to execute arbitrary system commands through the get_figma_data tool.
About this happening:
**figma-developer-mcp** has a **now-patched command injection** flaw, **CVE-2025-53967**, that can let attackers execute arbitrary commands and reach **remote code execution**. Th...
Figma-developer-mcp MCP server command injection (CVE-2025-53967)
VulnerabilityHow related: the "figma-developer-mcp" npm package has a bug (CVE-2025-53967; CVSS 7.5) that allows attackers to execute arbitrary system commands through the get_figma_data tool.
About this happening: **figma-developer-mcp** has a **now-patched command injection** flaw, **CVE-2025-53967**, that can let attackers execute arbitrary commands and reach **remote code execution**. Th...
Checkmarx Zero LITL prompt-injection analysis against Anthropic Claude Code
Technical Analysis
First: 15.09.2025 12:11
Last: 15.09.2025 12:11
Sources 1
About this happening:
Researchers demonstrated **lies-in-the-loop (LITL)**, a **prompt-injection** technique that can trick **Anthropic's Claude Code** into approving dangerous actions, expanding risk...
Checkmarx Zero LITL prompt-injection analysis against Anthropic Claude Code
Technical AnalysisAbout this happening: Researchers demonstrated **lies-in-the-loop (LITL)**, a **prompt-injection** technique that can trick **Anthropic's Claude Code** into approving dangerous actions, expanding risk...
Timeline
-
08.10.2025 20:14 1 articles · 7mo ago
CVE-2025-53967 disclosure in Figma MCP
Initial DisclosureResearchers identified a high-severity command injection flaw in the figma-developer-mcp npm package, tracked as CVE-2025-53967 and rated CVSS 7.5, where unvalidated user input reaches `child_process.exec()` through the `get_figma_data` tool and can enable arbitrary system commands, remote code execution, and system compromise in Figma MCP deployments.
Show sources
- Figma MCP Server Opens Orgs to Agentic AI Compromise — www.darkreading.com — 08.10.2025 20:14
-
08.10.2025 20:14 2 articles · 7mo ago
Figma MCP 0.6.3 remediation guidance
Mitigation Patch UpdateThe recommended remediation for Figma MCP deployments is to upgrade to version 0.6.3 or higher, replace `child_process.exec()` with `child_process.execFile()`, enforce input validation, audit systems still running vulnerable versions, and review logs for suspicious command execution patterns.
Show sources
- Figma MCP Server Opens Orgs to Agentic AI Compromise — www.darkreading.com — 08.10.2025 20:14
- Figma MCP Server Opens Orgs to Agentic AI Compromise — www.darkreading.com — 08.10.2025 20:14