debugger
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides structured instructions for root-cause analysis and debugging. It includes guardrails such as the 'honesty escape' to prevent hallucination and instructs the agent to rely only on observed evidence.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes external code and logs, though this is inherent to its debugging purpose. 1. Ingestion points: Code, logs, and bug reports (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: The skill instructions mention access to repo, shell, and test-execution tools. 4. Sanitization: Absent.
Audit Metadata