Fetching from the wire…
01
02
03
04
05
06
07
08
Top 5 · 2026-02-25 · source-backed
CVE-2025-59536 (CVSS 8.7): malicious hooks in a cloned repo's .claude/settings.json execute shell commands at session startup before security dialogs appear. MCP consent bypass: .mcp.json with enableAllProjectMcpServers auto-approves rogue servers. CVE-2026-21852 (CVSS 5.3): overriding ANTHROPIC_BASE_URL in project config redirects all API traffic — including plaintext auth headers — to attacker-controlled servers on startup with zero user interaction. All patched. Action: Never open untrusted repos without reviewing .claude/ directory contents. (Check Point Research, The Hacker News)
Each link below shares sources, entities, or timing with this story.
The most important security research this week. Check Point demonstrated three attack vectors in Claude Code exploiting project configuration files in untrusted repositories: (1) Hooks RCE (CVE-2025-59536, CVSS 8.7) — malicious hooks in .claude/settings.json execute shell comm...
Check Point Research disclosed that ANTHROPIC_BASE_URL in a repository's .claude config can redirect all API traffic — including full authorization headers — to an attacker-controlled server before the user reads a trust dialog, exfiltrating API keys in plaintext. A second vec...
Check Point Research disclosed that Claude Code's .claude/settings.json hooks can be weaponized in untrusted repos to execute arbitrary shell commands and exfiltrate Anthropic API keys by redirecting ANTHROPIC_BASE_URL to an attacker-controlled MitM proxy (Check Point). Both C...
Check Point Research disclosed CVE-2025-59536 and CVE-2026-21852 — two vulnerabilities that weaponize Claude Code's project configuration system against its users. This matters because an Agents Anonymous survey this week showed 90% of practitioners at their SF meetup use Clau...
An open-source AI attack platform called CyberStrikeAI, built in Go and integrating 100+ security tools with Claude and DeepSeek via a custom MCP server (ARXON), was used to systematically compromise 600+ Fortinet FortiGate appliances across 55 countries. Claude's coding agent...
The agent skills supply chain is under coordinated attack. Snyk's ToxicSkills audit found 36% of ClawHub's 3,984 skills contain prompt injection payloads, 13.4% have critical malware, and submission rates exploded 10x to 500+/day. This week alone: CVE-2026-2256 (CVSS 9.1) is a...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.