The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the @useclawlink/cli package from the NPM registry. This is a vendor-owned resource associated with the skill author.
[COMMAND_EXECUTION]: The instructions direct the agent to run several shell commands (npx @useclawlink/cli) to manage PagerDuty services, including listing actions and executing them.
[CREDENTIALS_UNSAFE]: The skill documents the local storage of credentials at ~/.clawlink/credentials.json as a result of the login process, providing a known path for sensitive data.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it interpolates user-supplied strings and JSON into shell commands that interact with external APIs.
Ingestion points: User-provided strings in the actions command and JSON payloads in the run command.
Boundary markers: None identified in the provided command examples.
Capability inventory: The skill allows for significant modifications to PagerDuty, such as creating triggers and managing service dependencies, via the CLI.
Sanitization: The documentation recommends a 'read first' approach and manual user confirmation before writing, but no technical sanitization or validation logic is defined within the skill itself.

pagerduty