dbs-deconstruct
Pass
Audited by Gen Agent Trust Hub on Mar 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's instructions are focused on providing a text-based analytical framework. No patterns of prompt injection intended to bypass safety filters or override agent constraints were detected.
- [DATA_EXFILTRATION]: No network requests, API calls, or access to sensitive local file paths (such as credentials or SSH keys) were identified. The skill only references its own local knowledge base and other skills by the same author.
- [REMOTE_CODE_EXECUTION]: The skill contains no executable code, script downloads, or package dependencies (Python or Node.js). All operations are performed via natural language processing within the agent's context.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided text for conceptual analysis (Phase 1). While it does not utilize explicit boundary markers to wrap untrusted input, it lacks the necessary capabilities (such as shell command execution or network access) that would be required to facilitate an exploitation of this surface.
Audit Metadata