The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the pscale CLI binary to perform database administrative tasks such as creating, updating, and testing webhooks. The reference documentation includes a SHA-256 checksum for binary verification.
[CREDENTIALS_UNSAFE]: The skill manages sensitive data including webhook signing secrets and HTTP Authorization headers. It implements defensive measures by instructing the agent to use jq for mandatory secret redaction and recommends using environment variables to avoid credential exposure in command history or logs.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The agent processes structured data returned from the PlanetScale API via CLI commands (list, show, create, update).
Boundary markers: None present; the agent relies on the structured nature of CLI output.
Capability inventory: The skill facilitates command execution (pscale) and network triggers (webhook test).
Sanitization: The skill provides explicit instructions to sanitize output by deleting secret fields using jq filters before the agent ingests the data.

pscale-webhook