Cognitive Firewall: A Proactive, Zero-Trust, Multi-Gate Framework for LLM Safety

Large language models LLMs can be induced to produce harmful content through multi turn strategies in which no single user message appears clearly unsafe. Existing runtime safeguards commonly evaluate prompts or responses as isolated messages, which limits their ability to recover ac-cumulated...

OpenClaw's Conflicting Tool Identity Hints Bypass Dangerous-Tool Prompting

Summary ACP permission resolution trusted conflicting tool identity hints from rawInput and metadata, which could suppress dangerous-tool prompting. Affected Packages / Versions - Package: openclaw npm - Affected: = 2026.3.22 - Latest released tag checked: v2026.3.23-2...