state-tax-estimator
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external transaction data which creates an ingestion surface for indirect prompt injection.
- Ingestion points: Financial data is retrieved from user accounts using the
spending_summaryandtransaction_searchtools as described in the workflow section of SKILL.md. - Boundary markers: The skill instructions do not provide explicit delimiters or instructions to the AI to ignore potential commands embedded within transaction descriptions or metadata.
- Capability inventory: The skill utilizes
export_transactionsfor data handling and relies on internal agent logic for tax liability estimation. - Sanitization: There are no explicit steps provided for sanitizing or validating the retrieved financial data before it is processed by the agent.
- [SAFE]: All identified tools (
spending_summary,transaction_search,export_transactions) are consistent with the skill's stated purpose of financial analysis and tax estimation. No malicious patterns, obfuscation, or unauthorized data exfiltration were detected.
Audit Metadata