OpenClaw publishes a batch of high-severity security advisories

Published a batch of high-severity security advisories — missing admin authorization on model overrides, plugin-install policy bypass, exposure of owner-only tools via MCP loopback, remote symlink following in mirror sync, and more

Why it matters

Reveals that the agent runtime's permission boundaries were imprecise on every front

Source — Primary sources first; secondary reports are labelled “Press”

Officialgithub.comhttps://github.com/openclaw/openclaw/security/advisories

Added in Week 35, 2026 · See that week

Same week

Figure 03 deployed in sequencing at Spartanburg Hall 52OfficialDiagnosing and Mitigating Context Rot in Long-horizon Search publishedPaperVoyage AI releases voyage-context-4Official

Nearby on the same axis

Agentic Misalignment in Summer 2026 (14 models)OfficialVerbalizable Representations Form a Global Workspace (J-space)PaperOpenAI and Hugging Face disclose a GPT-5.6 Sol evaluation security incidentOfficialAnthropic proposes an Advanced AI Framework for policyOfficial