ramp-core-workflow-b

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from the Ramp API (transactions, merchant names, and cardholder info) which could contain malicious instructions if the source is compromised.
  • Ingestion points: Fetches external transaction data using requests.get in SKILL.md.
  • Boundary markers: No specific delimiters or warnings are used when processing the transaction data.
  • Capability inventory: Includes network operations via the Python requests library and tool access for Bash(curl:*) and Bash(npm:*) in the YAML frontmatter.
  • Sanitization: The skill does not perform sanitization or validation on the fields returned by the API before processing or displaying them.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 10:51 PM
Security Audit — agent-trust-hub — ramp-core-workflow-b