ap-master

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill facilitates a sensitive financial process by ingesting data from external sources and configuration files without explicit safeguards.
  • Ingestion points: Data retrieved from [ERP] and [CASH] systems in Step 1 and Step 2, and the ../../CLAUDE.md configuration file.
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat external data as untrusted or to ignore instructions embedded within the data.
  • Capability inventory: The skill orchestrates high-impact actions including triple-matching, payment scheduling, and the generation of batch payment instructions for bank-enterprise direct connections.
  • Sanitization: No validation, escaping, or sanitization logic is present to handle potentially malicious input from the source systems.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 02:25 AM
Security Audit — agent-trust-hub — ap-master