kernel-scheduler-mm-vfs-internals
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides shell scripts and verification instructions that execute standard Linux system commands like cat, grep, and uname to inspect kernel state through the /proc filesystem. It also includes commands to enable and configure kernel event tracing via the /sys/kernel/tracing/ interface, which typically requires elevated privileges.
- [INDIRECT_PROMPT_INJECTION]: The skill reads and processes output from system-level files and trace logs, establishing an attack surface where maliciously crafted system telemetry could potentially influence the agent's reasoning. The skill lacks explicit boundary markers or sanitization for this ingested data. Ingestion points: Diagnostic scripts in the examples/good/ directory that read from /proc and /sys files. Boundary markers: Absent. Capability inventory: Read operations and kernel tracing configuration. Sanitization: Absent.
Audit Metadata