The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references multiple official documentation pages and API schemas hosted on developers.cloudflare.com. These are official resources from a well-known service provider.
[INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection because it is designed to ingest and act upon data from external and account-specific sources. * Ingestion points: The skill retrieves information from developers.cloudflare.com, Cloudflare API schemas, and existing account resources such as Access policies, Gateway rules, and device profiles. * Boundary markers: There are no specific instructions for the agent to use delimiters or ignore potentially malicious instructions embedded within the retrieved data. * Capability inventory: The agent is authorized to use MCP tools to inspect, create, and modify security policies and network configurations. * Sanitization: The skill does not outline procedures for sanitizing or validating external data before it is processed or used in configuration tasks.

cloudflare-one