quantum-debugger

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill references several powerful system diagnostic and debugging utilities, including gdb, lldb, valgrind, perf, strace, ltrace, and bpftrace. These tools are standard for software engineering and debugging but require significant system permissions and can have substantial performance or security impacts if misused.
  • [PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data, creating an indirect prompt injection attack surface.
  • Ingestion points: Collects error messages, stack traces, and application logs during the 'Observe' phase (SKILL.md).
  • Boundary markers: No instructions are provided for isolating untrusted log content with delimiters or specific ignore warnings.
  • Capability inventory: The skill instructs the agent to use invasive system tools like gdb, perf, and bpftrace to analyze behavior (SKILL.md).
  • Sanitization: There is no description of filtering or validating the external log data before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 01:56 AM
Security Audit — agent-trust-hub — quantum-debugger