AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

cloudflare

Discover and choose Cloudflare products for apps, APIs, AI agents, storage, networking, and security. Use for architecture and product selection, including when the user describes a need without naming a Cloudflare product; then find the relevant skill or documentation.

Original source
cloudflare
License
Apache-2.0
Pinned revision
e44f580f9c4fa3bf268e284ae7706c7954ba31fd

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

Ask your assistant

Use Agentic Memory skills to read cloudflare--cloudflare at revision e44f580f9c4fa3bf268e284ae7706c7954ba31fd, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 135

references/workers-ai/gotchas.md

Workers AI Troubleshooting

Use the actual error, model identifier, integration, and installed versions to choose the relevant reference.

| Symptom or decision | Documentation and checks | |---------------------|--------------------------| | Missing binding or types | Binding configuration and Workers TypeScript; check the environment being run | | Development inference fails | Workers and Wrangler setup; check account access and binding setup | | Unknown model, invalid input, or unexpected response | Open the exact model in the catalog; check its schema, context window, and feature support | | Inference error or retry decision | Error codes and HTTP statuses | | Throttling or concurrency planning | Current limits | | Usage or cost estimate | Current pricing; use the selected model's billing units and expected workload | | Old SDK examples fail | Native binding, AI SDK, or OpenAI compatibility, according to the integration |

Do not copy an error-code mapping, per-request neuron estimate, or context-window range from another model or an older example. Measure latency for the intended workload rather than promising a fixed cold-start or inference time.