The Agent Skills Directory

[COMMAND_EXECUTION]: The script interacts with the network using subprocess.check_output to execute curl. This is implemented safely by passing a list of arguments rather than a raw shell string, which prevents command injection from user-supplied inputs such as usernames or tweet IDs.
[DATA_EXPOSURE]: The skill manages its API key by storing it in a local configuration file at ~/.config/twitterapi-io/config.json. The implementation uses chmod_600 to ensure that the file has restrictive permissions (read/write only by the owner), which is a standard security best practice for CLI tools handling sensitive credentials.
[INDIRECT_PROMPT_INJECTION]: Because the skill fetches and processes data from external, untrusted sources (Twitter), it inherently possesses an indirect prompt injection surface. However, the risk is mitigated as the skill only performs read-only operations and returns structured JSON data, rather than executing commands based on the fetched content.

twitterapi-io