OpenClaw, an open-source AI agent project, has significantly hardened its security after receiving 1,309 security advisories since January 10, with 87% of critical reports found to be invalid. The pro…
OpenClaw launched Skill Workshop, a tool that requires agents to submit proposed reusable procedures as reviewable drafts before they become active skills. The system prevents agents from directly wri…
OpenClaw has partnered with NVIDIA to strengthen security for AI agent skill files, integrating NVIDIA's SkillSpector scanner and Skill Card trust artifacts into its ClawHub registry. The collaboratio…
OpenClaw has introduced a new "auto" approval mode for host command execution, designed to balance safety and efficiency by allowing low-risk commands to run automatically while routing uncertain or p…
OpenClaw's latest release, version 5.28, achieved a 5.1x speed increase in agent turns and reduced its published tarball size by 59% to 17.9 MB, while also cutting fresh install size by 52.8%. The per…
OpenClaw Security is developing new protections for its AI personal assistant platform, including filesystem boundary controls called fs-safe and a network egress routing layer called Proxyline. The p…
OpenAI models now run agent turns through the native Codex app-server harness by default in OpenClaw, replacing the previous approach where OpenClaw drove the model loop itself. The change gives Codex…