The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the @useclawlink/cli package from the NPM registry. This tool is required for authentication and interacting with DigitalOcean.
[COMMAND_EXECUTION]: The agent is instructed to run shell commands via the ClawLink CLI to perform actions such as listing, describing, and running DigitalOcean operations.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes data from external DigitalOcean resources. Malicious content within these resources (e.g., resource names or metadata) could potentially be interpreted as instructions by the agent.
Ingestion points: Data retrieved from DigitalOcean through the actions and describe commands in SKILL.md.
Boundary markers: There are no specific delimiters or instructions provided to the agent to treat external data as untrusted, although the skill recommends manual user confirmation for write operations.
Capability inventory: The agent has the ability to execute API calls and manage cloud infrastructure via the run command.
Sanitization: No explicit sanitization or filtering of external data is defined in the skill instructions.

digitalocean