The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a local script scripts/hashport.sh to calculate stable port numbers. The script is self-contained and uses standard system utilities (bash, printf, lsof) to perform its task. Findings show the script is robustly quoted to prevent injection from project names.
[DATA_EXPOSURE]: The skill instructs the agent to read project directory names and update configuration files like .env, package.json, and vite.config.ts. This access is restricted to standard project configuration management and aligns with the skill's stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill has a minor surface for indirect prompt injection as it processes project names and configuration file content. However, the logic for generating the port is algorithmic (converting strings to integers), and the capability involves writing standard configuration strings, which poses minimal risk. The script effectively sanitizes input by extracting individual characters and converting them to ASCII codes.

lov-project-port