The Agent Skills Directory

[SAFE]: The skill implements its core charting functionality using reputable libraries and follows established patterns for CLI tools. No malicious behavior, obfuscation, or unauthorized data collection was detected.\n- [COMMAND_EXECUTION]: The scripts/chart.mjs script is called via the command line. Potential command injection risks are mitigated by the CAPABILITY.md file, which explicitly directs the agent to use structured process execution and avoid shell-based string interpolation.\n- [PROMPT_INJECTION]: The skill processes untrusted data for visualization. Ingestion points: Data is received through CLI flags in scripts/chart.mjs. Boundary markers: Safety instructions in CAPABILITY.md define secure interaction patterns. Capability inventory: The script performs file read/write operations for data input and image output. Sanitization: Input data is parsed through standard JSON and CSV parsers.

chart-image