The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches installation and configuration scripts from the author's official GitHub repositories and the 'getalby.com' domain.
[REMOTE_CODE_EXECUTION]: Orchestrates the execution of remote shell scripts using the curl | bash pattern. These scripts are hosted on the vendor's verified infrastructure.
[COMMAND_EXECUTION]:
Frequently executes the @getalby/hub-cli utility via npx for wallet management.
Performs system administration tasks including systemctl for service management and apt-get for installing dependencies.
Uses sudo for elevated operations such as editing systemd units and installing system packages.
Utilizes shell piping (e.g., echo | qrencode), which presents a potential command injection surface if the input strings from the lightning network are not properly sanitized.
[PROMPT_INJECTION]: Identifies a surface for indirect prompt injection through the ingestion of data from the lightning node.
Ingestion points: Processes untrusted external content such as transaction memos and application names from the output of list-transactions and list-apps.
Boundary markers: No specific delimiters or safety instructions are used to isolate this data from the agent's operational context.
Capability inventory: The agent has significant capabilities including shell execution, network access via the CLI, and local file system access.
Sanitization: No explicit sanitization or validation mechanisms are defined for data retrieved from external sources.

alby-hub-skill