netlify-ai-gateway
Reference for Netlify AI Gateway — the managed proxy that routes calls to OpenAI, Anthropic, and Google Gemini SDKs without provider API keys. Use this skill any time the user wants to add AI on a Netlify site (chat, completion, reasoning, image generation, image-to-image edit/stylize), choose or change a model, wire up the OpenAI / Anthropic / @google/genai SDK, decide which provider to use for an image-gen feature (it's Gemini-only on the gateway), or debug "model not found" / "API key missing" against the gateway. Required reading before pinning a model — the gateway exposes a curated subset, not every provider model.
68
82%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly defines a specific domain (Netlify AI Gateway), lists concrete capabilities and use cases, and provides explicit trigger conditions. It uses third person voice appropriately, includes natural keywords users would search for, and even includes helpful disambiguation notes (e.g., Gemini-only for image-gen, curated subset of models). The description is information-dense without being padded.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: routing calls to specific providers, adding AI features (chat, completion, reasoning, image generation, image-to-image edit/stylize), wiring up specific SDKs, choosing models, debugging specific errors like 'model not found' / 'API key missing'. | 3 / 3 |
Completeness | Clearly answers both 'what' (managed proxy that routes calls to AI providers without API keys, curated model subset) and 'when' with explicit triggers ('Use this skill any time the user wants to add AI on a Netlify site...choose or change a model...debug model not found...'). The 'Use this skill any time' clause is comprehensive and well-structured. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: 'Netlify', 'AI', 'OpenAI', 'Anthropic', 'Google Gemini', 'chat', 'completion', 'image generation', 'model not found', 'API key missing', 'SDK', specific SDK names like '@google/genai'. These are terms users would naturally use when encountering these scenarios. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — scoped specifically to Netlify AI Gateway, a very specific product/feature. The combination of Netlify + AI Gateway + specific provider SDKs + specific error messages creates a clear niche that is unlikely to conflict with generic AI or generic Netlify skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, highly actionable skill with excellent executable code examples for all three provider SDKs, image generation, and Netlify Functions integration. Its main weaknesses are moderate verbosity from repeated information (first-deploy requirement stated three times) and a lack of explicit validation/verification steps in the workflow. The models list is valuable reference material but contributes to the file's length — it could benefit from being a separate reference file.
Suggestions
Add a quick validation step after setup (e.g., a curl command or minimal script to confirm the gateway is active and env vars are injected correctly).
Consolidate the first-deploy requirement into a single prominent callout rather than repeating it in the callout, setup section, and local development section.
Consider extracting the Available Models list into a separate MODELS.md reference file to keep the main skill leaner and make model updates easier to manage.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is mostly efficient with good code examples and tables, but includes some unnecessary explanation (e.g., explaining what a proxy is, restating the first-deploy requirement three times across the callout, setup section, and local development section, and the env var table over-explains placeholder keys). The available models list is necessarily long but well-organized. | 2 / 3 |
Actionability | Excellent actionability — every SDK integration has complete, copy-paste-ready TypeScript code with install commands. The Netlify Function example is fully executable with config export. Image generation covers both text-to-image and image-to-image with concrete response parsing code. | 3 / 3 |
Workflow Clarity | The setup steps are clear (enable AI → deploy → install SDK), and troubleshooting covers common failure modes. However, there are no explicit validation checkpoints — for instance, no step to verify the gateway is active after first deploy, no way to confirm env vars are injected correctly, and no feedback loop for debugging model availability issues. | 2 / 3 |
Progressive Disclosure | The content is well-structured with clear sections and a logical flow from setup through providers to troubleshooting. However, the available models list (~40 lines) and the image generation section could be split into separate reference files. The single cross-reference to `netlify-blobs/SKILL.md` is well-placed, but the document is quite long for a single file with no bundle support files. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- netlify/context-and-tools
- Commit
cffaf74
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.