strace-ltrace
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- COMMAND_EXECUTION (SAFE): The skill provides numerous command-line examples for powerful system diagnostic tools (strace, ltrace, gdb). These tools are used for their intended primary purpose of debugging and do not involve unauthorized execution or malicious script downloads.
- INDIRECT_PROMPT_INJECTION (LOW): The skill involves reading and interpreting output from potentially untrusted binaries being traced. 1. Ingestion points: Terminal output from strace, ltrace, and grep. 2. Boundary markers: Absent; raw output is processed directly by the agent. 3. Capability inventory: Includes system-level tracing, process attachment, and debuggers. 4. Sanitization: Absent; the agent is expected to interpret raw diagnostic data which could theoretically be manipulated by a malicious binary under trace.
Audit Metadata