CVE-2026-35651

OpenClaw versions 2026.2.13 through 2026.3.24 contain an ANSI escape sequence injection vulnerability in approval prompts that allows attackers to spoof terminal output. Untrusted tool metadata can carry ANSI control sequences into approval prompts and permission logs, enabling attackers to...

CVE-2026-35651

OpenClaw OpenClaw 2026.2.13 through 2026.3.24 contains an ANSI escape sequence injection vulnerability in approval prompts. Untrusted tool metadata can carry ANSI control sequences into approval prompts and permission logs, enabling attackers to spoof terminal output by manipulating displayed inf...

PT-2026-31962

OpenClaw versions 2026.2.13 through 2026.3.24 contain an ANSI escape sequence injection vulnerability in approval prompts that allows attackers to spoof terminal output. Untrusted tool metadata can carry ANSI control sequences into approval prompts and permission logs, enabling attackers to...

CVE-2026-32898 OpenClaw < 2026.2.23 - ACP Permission Auto-Approval Bypass via Untrusted Tool Metadata

OpenClaw versions prior to 2026.2.23 contain an authorization bypass vulnerability in the ACP client that auto-approves tool calls based on untrusted toolCall.kind metadata and permissive name heuristics. Attackers can bypass interactive approval prompts for read-class operations by spoofing tool...

EUVD-2026-13976

OpenClaw versions prior to 2026.2.23 contain an authorization bypass vulnerability in the ACP client that auto-approves tool calls based on untrusted toolCall.kind metadata and permissive name heuristics. Attackers can bypass interactive approval prompts for read-class operations by spoofing tool...

CVE-2026-32898

CVE-2026-32898 affects OpenClaw versions prior to 2026.2.23, specifically the ACP client. The issue is an authorization bypass where tool calls are auto-approved based on untrusted toolCall.kind metadata and permissive name heuristics. Attackers can bypass interactive approval prompts for read-cl...

Attractive Metadata Attack: Inducing LLM Agents to Invoke Malicious Tools

Large language model LLM agents have demonstrated remarkable capabilities in complex reasoning and decision-making by leveraging external tools. However, this tool-centric paradigm introduces a previously underexplored attack surface: adversaries can manipulate tool metadata -- such as names,...