margin-analyzer

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection Surface: The skill allows users to upload transaction and cost data via CSV files as a fallback mechanism. This creates a potential surface where malicious instructions could be embedded within the data fields to influence the agent's behavior during analysis.
  • Ingestion points: Data entry occurs through CSV uploads defined in SKILL.md (Step 1 and Step 4) and specified in the schema at reference/csv-schema.md.
  • Boundary markers: The instructions do not specify the use of delimiters or clear "ignore embedded instructions" warnings for the agent when processing the content of these files.
  • Capability inventory: The skill possesses the capability to interact with external business APIs via tools such as profit-loss-quickbooks-account, list_transactions, and make_api_request.
  • Sanitization: While reference/csv-schema.md discusses sanitizing data for formatting purposes (e.g., stripping currency symbols), there are no specific measures described for sanitizing the input against prompt injection attempts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 11:27 PM
Security Audit — agent-trust-hub — margin-analyzer