The Agent Skills Directory

[Data Ingestion Consideration]: The skill describes building Workers that process external HTTP request data (via the request object) and identity context. In the context of middleware development, these inputs are untrusted and represent a surface for indirect prompt injection or logic manipulation. The instructions appropriately suggest best practices, such as separate error handling for identity lookups and guarding optional posture values.
[Network and Continuation Capabilities]: The instructions detail the use of fetch() for subrequests and next() for request continuation. While these are powerful capabilities that allow the Worker to interact with external origins, they are core features of the Cloudflare Workers platform. The skill provides clear guidance on the limitations of these tools, such as how fetch() creates independent subrequests that do not inherit the user's Gateway policies.
[Official Vendor Documentation]: The skill makes extensive use of links to developers.cloudflare.com to provide necessary implementation details. These references are used to ensure the developer uses the most current API payloads and limit configurations from the official vendor source.

gateway-workers