debug-like-expert
Pass
Audited by Gen Agent Trust Hub on Mar 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses non-privileged shell commands like
find,grep, andlsto identify project file types and locate expertise documentation within the user's local directory (~/.claude/skills/expertise/). These operations are limited to discovery tasks to provide context-aware debugging advice. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its primary purpose is to read and analyze untrusted source code and external research results.
- Ingestion points: SKILL.md (via project file discovery and reading operations).
- Boundary markers: None identified in the skill's instructions.
- Capability inventory: Directory enumeration and file reading.
- Sanitization: No explicit sanitization or filtering of the analyzed file content or search results is implemented.
Audit Metadata