Name: ai-gateway
Rating: 91.4 (1 reviews)
Author: vercel

ai-gateway

Vercel AI Gateway guidance for setup, model discovery, authentication, routing, fallbacks, BYOK, budgets, spend reporting, observability, compatible APIs, and coding-agent configuration. Use when adding AI Gateway to an app, migrating provider calls, choosing models or providers, debugging gateway requests, or running `vercel ai-gateway` commands.

1.08x

Quality

91%

Does it follow best practices?

Impact

93%

1.08x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Quality

Content

96%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Exemplary skill body: executable code and commands, a sequenced workflow with explicit validation and an authorization gate, and a clean overview-plus-references structure where all four linked bundle files exist. The only tightening opportunity is removing the small amount of repeated 'fetch live models, don't trust memory' guidance and pruning the link list.

Dimension	Reasoning	Score
Conciseness	The body is dense and table-driven with no tutorials on concepts Claude already knows — every section carries operational facts (exact endpoints, env var names, CLI subcommands), and time-sensitive details like 'Node.js 22 or later' are properly hedged. It falls just short of the lean anchor-5 due to minor repetition (the fetch-/v1/models admonition appears in both the sources and model sections) and a 10-link documentation dump that could be trimmed.	4 / 5
Actionability	Guidance is fully executable throughout: a runnable `curl -fsSL https://ai-gateway.vercel.sh/v1/models` command, a copy-paste-ready `generateText` TypeScript snippet with env validation, a concrete CLI inventory table, exact `baseURL` values per existing SDK, and named env vars (`AI_GATEWAY_API_KEY`, `VERCEL_OIDC_TOKEN`) covering the common integration cases.	5 / 5
Workflow Clarity	The 'Implementation workflow' gives a clear 9-step sequence with explicit validation checkpoints — run formatter/type checker/focused tests, make one authorized live request, verify in Gateway Logs — plus a verification checklist and gating language for risky outward-facing actions (spend credits, create keys, change budgets). Error-recovery guidance ('Common gateway outcomes include authentication failure, insufficient credits...') is present, matching the anchor-5 pattern.	5 / 5
Progressive Disclosure	The body is a well-organized overview that routes to four real one-level-deep bundle files via the 'Route the request to the right guide' table and inline links (setup.md, routing.md, spend-observability.md, coding-agents.md — all verified present in ./references/), with detail correctly deferred (e.g. 'Read references/routing.md before adding any of these fields').	5 / 5
	Total	19 / 20 Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: it names the product and its full capability surface, and pairs a clear what with an explicit multi-trigger Use-when clause in third-person voice. The only refinement left is replacing the noun-list of topics with more concrete action verbs and adding a few common synonyms to the trigger set.

Dimension	Reasoning	Score
Specificity	The description enumerates many concrete capability areas — 'setup, model discovery, authentication, routing, fallbacks, BYOK, budgets, spend reporting, observability, compatible APIs, and coding-agent configuration' — giving comprehensive domain coverage. It stays a notch below a 5 because these are noun-listed topics under generic 'guidance' rather than sharp concrete action verbs ('Extract text', 'fill forms').	4 / 5
Completeness	It explicitly answers both questions: the first sentence states what the skill provides (the enumerated capability areas), and 'Use when adding AI Gateway to an app, migrating provider calls... debugging gateway requests, or running `vercel ai-gateway` commands' supplies concrete trigger phrases for when to use it.	5 / 5
Trigger Term Quality	The 'Use when' clause covers natural phrasings users would actually say: 'adding AI Gateway to an app, migrating provider calls, choosing models or providers, debugging gateway requests, or running `vercel ai-gateway` commands'. A few natural synonyms are missing (e.g. 'cost tracking', 'failover', 'gateway key'), keeping it just below the comprehensive anchor-5 coverage.	4 / 5
Distinctiveness Conflict Risk	It is anchored to the specific named product 'Vercel AI Gateway' with product-unique triggers (BYOK, gateway budgets, `vercel ai-gateway` commands), carving a clear niche with minimal risk of firing for unrelated gateway, proxy, or SDK skills.	5 / 5
	Total	18 / 20 Passed

Validation

81%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: vercel/vercel-plugin
Path: skills/ai-gateway/SKILL.md
Commit: c632a50

Reviewed: 1 hour ago

Table of Contents

Discovery Implementation Validation

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.