The Agent Skills Directory

[SAFE]: The skill implements music generation functionality as described. All operations, including task submission and status polling, are performed through official vendor-controlled scripts and domains.
[DATA_EXFILTRATION]: Network activity is restricted to the legitimate API endpoint https://giggle.pro. No unauthorized data exfiltration or access to sensitive local files was observed. The skill correctly instructs the agent to use environment variables for the API key (GIGGLE_API_KEY) and the script implementation uses os.getenv instead of hardcoding secrets.
[COMMAND_EXECUTION]: The skill executes a local Python script provided with the package. Arguments passed to the script, such as prompts or style, are handled safely via standard command-line argument parsing and JSON encoding in network requests, preventing command injection vulnerabilities.

giggle-generation-music