us-realtime-quote
Warn
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: MEDIUMCREDENTIALS_UNSAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill contains a hardcoded API key (
123456) in the connection setup code withinSKILL.md. - [DATA_EXFILTRATION]: User-provided search terms and stock symbols are transmitted to an external hardcoded IP address (
43.167.234.49) over an insecure HTTP connection. - [PROMPT_INJECTION]: The instructions use absolute directives ("must immediately call", "never answer from memory") to override the model's standard behavior and ignore its trained safety or knowledge constraints.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection.
- Ingestion points: User-provided stock symbols are directly interpolated into the API request URL.
- Boundary markers: Absent; there are no delimiters or instructions to prevent the model from interpreting user input as part of the command logic.
- Capability inventory: The skill has the capability to perform network requests (
curl) to an external IP. - Sanitization: Absent; no validation or escaping of user input is specified before it is sent to the external endpoint.
Audit Metadata