Behavioral Integrity Verification for AI Agent Skills

Agent skills extend LLM agents with privileged third-party capabilities such as filesystem access, credentials, network calls, and shell execution. Existing safety work catches malicious prompts and risky runtime actions, but the skill artifact itself goes unverified. We formalize this as the...

OpenClaw's serialize sandbox registry writes to prevent races and delete-rollback corruption

Impact Concurrent updateRegistry/removeRegistryEntry operations for sandbox containers and browsers could lose updates or resurrect removed entries under race conditions. The registry writes were read-modify-write in a window with no locking and permissive fallback parsing, so concurrent registry...