Security analysis of OpenClaw reveals composable RCE paths from LLM tool calls, invalid closed-world assumptions in exec allowlists, and plugin-based attacks that bypass runtime policy.
Maliciousyahoofinanceskill distributing two-stage dropper viaclawhub.ai, February 2026.https://github.com/openclaw/openclaw/issues/ 5675
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Security Analysis of the OpenClaw AI Agent Framework
Security analysis of OpenClaw reveals composable RCE paths from LLM tool calls, invalid closed-world assumptions in exec allowlists, and plugin-based attacks that bypass runtime policy.