chart-visualization
Pass
Audited by Gen Agent Trust Hub on Apr 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Node.js script (
scripts/generate.js) to handle chart generation. This script is designed to accept either a raw JSON payload or a file path as its primary command-line argument. - [DATA_EXFILTRATION]: The
scripts/generate.jsscript transmits chart data to an external service endpoint atantv-studio.alipay.com. This is the official API for the AntV visualization suite provided by Ant Group (Alipay), a well-known technology provider. - [PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection because it processes untrusted user-supplied data and has the capability to send that data to a remote server.
- Ingestion points: Data for various chart types (26 options) is extracted from user inputs as described in
SKILL.mdand thereferences/directory. - Boundary markers: None; the skill instructions do not specify any delimiters or safety warnings to ignore instructions that might be embedded within the user data.
- Capability inventory: The skill can perform network operations via
fetchto the AntV API and read local files viafs.readFileSyncwithin thescripts/generate.jsfile. - Sanitization: None; the input data is parsed as JSON and passed directly to the external API without specific sanitization or validation against embedded natural language instructions.
Audit Metadata