The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill facilitates the installation of the ralpher-cli tool by executing a remote script directly in the shell: curl -fsSL https://raw.githubusercontent.com/pablozaiden/ralpher/main/install.sh | sh. This script is hosted on the skill author's GitHub repository and represents the standard installation procedure for the service.
[EXTERNAL_DOWNLOADS]: The skill references and downloads the installation script and future updates from the author's GitHub infrastructure (github.com/pablozaiden).
[COMMAND_EXECUTION]: The skill is centered around executing shell commands to interact with a Ralpher server, including discovering APIs, checking status, and managing websocket streams.
[PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection due to its handling of data from external sources.
Ingestion points: Remote data enters the system through API responses fetched by ralpher-cli api and event streams from ralpher-cli ws (SKILL.md, REFERENCE.md).
Boundary markers: There are no specific delimiters or instructions present to prevent the agent from following directions potentially embedded in external data.
Capability inventory: The skill allows for shell command execution and network requests via the Ralpher CLI tool.
Sanitization: The instructions indicate that the tool validates JSON payloads, but no specific sanitization of retrieved data content is mentioned.

ralpher-cli