code-refactoring-context-restore
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions and code snippets are entirely consistent with the stated purpose of assisting in code refactoring and context management. No evidence of prompt injection, data exfiltration, or malicious command execution was found.
- [PROMPT_INJECTION]: The skill facilitates the ingestion of external project context via the
context_sourceparameter. While this creates an inherent surface for indirect prompt injection, the skill emphasizes relevance filtering and semantic ranking, and no active injection patterns are present. - Ingestion points:
context_sourceparameter inSKILL.md. - Boundary markers: None explicitly defined.
- Capability inventory: Reasoning based on restored context and use of the
context-restoreCLI tool. - Sanitization: Implements semantic similarity thresholds and relevance-based ranking logic.
Audit Metadata