systematic-debugging
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a systematic root cause analysis framework focused on evidence-based debugging. It includes proactive security measures, such as recommending the use of isolated sandboxes for test execution and enforcing strict verification gates to prevent unverified code changes.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes external diagnostic data. 1. Ingestion points: Log files, stack traces, and execution inputs (SKILL.md). 2. Boundary markers: Structured output contracts like test-report.json are used for artifacts. 3. Capability inventory: File manipulation and command execution tools (SKILL.md). 4. Sanitization: Instructions mandate execution of reproduction scripts in isolated sandboxes with restricted network access and non-root users to mitigate risks.
Audit Metadata