seoul-subway-arrival
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
curlutility within its workflow to fetch data from an external API. While used here for transit data, the execution of shell-based networking tools allows the agent to interact with arbitrary endpoints if configured with a malicious URL. - [EXTERNAL_DOWNLOADS]: The skill relies on an external proxy server, with the address determined by the
KSKILL_PROXY_BASE_URLenvironment variable. This introduces a dependency on external infrastructure that is not under the direct control of the skill or the local environment. - [PROMPT_INJECTION]: The skill processes and summarizes data from an external API response, creating a surface for indirect prompt injection. If the proxy or the upstream data source were compromised, it could return instructions designed to influence the agent's behavior during the summarization phase.
- Ingestion points: External API response from the URL defined in
KSKILL_PROXY_BASE_URL(SKILL.md). - Boundary markers: Absent from the instruction set.
- Capability inventory: Shell command execution via
curl(SKILL.md). - Sanitization: No explicit sanitization or validation of the API response is described before the data is summarized by the agent.
Audit Metadata