ramp-hello-world
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill includes benign documentation and implementation details for accessing financial data through the Ramp API. The code snippets perform standard data retrieval and display tasks without evidence of malicious intent or hidden functionality.- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external data from the Ramp API, which creates a potential surface for indirect prompt injection if the source data is manipulated.
- Ingestion points: API response data retrieved in
SKILL.md(e.g., card names, transaction descriptions, user details). - Boundary markers: None present in the provided code snippets to delimit untrusted API data.
- Capability inventory: The skill allows access to file system tools (Read, Write, Edit) and shell commands (Bash, Grep).
- Sanitization: No data sanitization or input validation is implemented in the instructional snippets.
Audit Metadata