tax-compliance-verification
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWPROMPT_INJECTIONNO_CODE
Full Analysis
- [NO_CODE] (SAFE): The skill consists entirely of markdown instructions, tables, and reporting templates. No scripts, binaries, or automated execution steps are included.
- [PROMPT_INJECTION] (LOW): This finding identifies an Indirect Prompt Injection surface. 1. Ingestion points: The skill instructs the agent to read and verify code files in lib/analysis/, lib/ai/, and app/api/audit/. 2. Boundary markers: The instructions lack delimiters to separate untrusted code from agent instructions. 3. Capability inventory: The agent is limited to generating text-based XML reports; it possesses no file-write, network-access, or command-execution capabilities. 4. Sanitization: No sanitization or validation of the audited content is performed.
Audit Metadata