The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands to detect the operating system, check for Node.js and OpenClaw prerequisites, and manage service restarts via openclaw gateway restart.
[REMOTE_CODE_EXECUTION]: The skill uses npx -y openclaw-openviking-setup-helper@latest to download and execute a setup script from the npm registry. This is a standard installation pattern for the OpenViking ecosystem and targets a well-known package registry.
[EXTERNAL_DOWNLOADS]: The skill performs downloads from clawhub (the primary plugin registry) and npm. These are well-known services used for distributing software packages. It also performs connectivity checks to a user-provided BASE_URL using curl or Invoke-WebRequest.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data, such as the server URL and API keys, and interpolates them into shell command flags. While this creates a surface for command injection, the instructions include specific normalization steps (e.g., protocol checking for URLs) and regex validation for identifiers like PEER_PREFIX to mitigate risks.
[PROMPT_INJECTION]: The instructions contain directives for the agent to perform certain steps 'silently' or to 'not show this step to the user.' In the context of this skill, these are benign interaction design patterns intended to manage the user interface and do not indicate malicious concealment.

install-openviking-memory