The Agent Skills Directory

[DATA_EXFILTRATION]: The skill manages a sensitive API token stored at ~/.config/cloudflare/token. While this constitutes a credential storage surface, the save_token.rb script correctly implements secure file permissions (0600) to restrict access to the current user.
[DATA_EXFILTRATION]: Network operations are directed exclusively to Cloudflare's official API endpoint (api.cloudflare.com). These communications are performed over HTTPS using Ruby's standard library.
[COMMAND_EXECUTION]: The skill executes local Ruby scripts to interact with the API. All scripts are part of the distributed skill package and do not utilize external dependencies or gems, reducing the risk of supply chain attacks.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources (the Cloudflare API) that may contain content controlled by other parties (e.g., DNS record names, Worker script IDs).
Ingestion points: Data is fetched in scripts/zones.rb, scripts/zone.rb, scripts/dns_records.rb, scripts/workers.rb, and scripts/pages_projects.rb.
Boundary markers: The output is presented directly to the user/agent without specific delimiters.
Capability inventory: The skill has the capability to perform state-changing HTTP requests (POST, PUT, DELETE, PATCH) as defined in scripts/auth.rb.
Sanitization: There is no explicit sanitization of the API responses before they are printed to the console for the agent to process.

cloudflare