The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard command-line tools (ffmpeg, ffprobe, buzz) for media processing and message delivery. These are well-defined and scoped to the skill's stated purpose of attaching media to a specific platform.- [CREDENTIALS_SAFE]: The skill explicitly demonstrates safe credential handling. In step 4, it uses python-dotenv to parse an environment file and injects values directly into a subprocess environment rather than sourcing them in a shell, which prevents accidental credential exposure or shell-mangling of complex tokens. It also contains explicit instructions ('Pitfalls' section) to never print private keys or environment variables.- [DATA_EXPOSURE]: The skill accesses a local log file (~/.hermes/logs/gateway.log) to retrieve a channel ID. This is a legitimate operational requirement for the tool to function and does not involve exfiltrating sensitive data to unauthorized third parties.- [REMOTE_CODE_EXECUTION]: All execution is performed locally using installed CLI tools. There are no patterns of downloading and executing remote scripts or unverified dependencies.

buzz-media-attachments