CtrlK
BlogDocsLog inGet started
Tessl Logo

ai-sdk

Answer questions about the AI SDK and help build AI-powered features. Use when developers: (1) Ask about AI SDK functions like generateText, streamText, ToolLoopAgent, embed, or tools, (2) Want to build AI agents, chatbots, RAG systems, or text generation features, (3) Have questions about AI providers (OpenAI, Anthropic, Google, etc.), streaming, tool calling, structured output, or embeddings, (4) Use React hooks like useChat or useCompletion. Triggers on: "AI SDK", "Vercel AI SDK", "generateText", "streamText", "add AI to my app", "build an agent", "tool calling", "structured output", "useChat".

88

2.92x
Quality

89%

Does it follow best practices?

Impact

76%

2.92x

Average score across 3 eval scenarios

SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is ai-sdk in vercel/ai

SKILL.md
Quality
Evals
Security

Quality

Content

78%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.

A well-organized operational skill body that tells Claude how to find current AI SDK documentation rather than trusting memory, with concrete commands, a typecheck validation loop, and clean progressive disclosure into four reference files. It is slightly verbose in places and defers most SDK usage code to the bundled docs.

Suggestions

Tighten the "Critical: Do Not Trust Internal Knowledge" preamble and point 7; the warning framing and the long curl explanation can be shortened without losing the instruction.

Add one minimal inline runnable example (e.g. a generateText or ToolLoopAgent snippet) so common cases are copy-paste ready without first grepping the docs.

Consolidate the version-split "Finding Documentation" guidance so the ai@6.0.34+ and earlier-version paths are presented as a single decision tree, reducing redundant section switching.

DimensionReasoningScore

Conciseness

Mostly lean and directive without explaining concepts Claude already knows, but the opening "Everything you know about the AI SDK is outdated or wrong" and the lengthy model-fetching curl note in point 7 could be trimmed slightly.

4 / 5

Actionability

Provides concrete executable commands (pnpm add ai, grep paths, the curl+jq model-ID command), but defers actual SDK usage code to bundled docs/source rather than showing a runnable generateText or agent example inline.

4 / 5

Workflow Clarity

Clear sequence (install ai -> search bundled docs -> search web -> typecheck) with an explicit validation checkpoint (run typecheck) and a feedback loop for typecheck failures via common-errors.md, with only minor gaps.

4 / 5

Progressive Disclosure

Body is an overview that points to four real one-level-deep reference files (common-errors.md, ai-gateway.md, type-safe-agents.md, devtools.md), each clearly signaled with a description in the References section, making navigation easy.

5 / 5

Total

17

/

20

Passed

Description

100%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 well-constructed description that pairs a concise capability statement with an enumerated use-when list and an explicit trigger-term block. It answers both what and when, uses third-person/imperative voice, and is tightly scoped to the AI SDK niche.

DimensionReasoningScore

Specificity

Names concrete actions ("Answer questions", "help build AI-powered features") plus an enumerated list of specific functions (generateText, streamText, ToolLoopAgent, embed) and feature types (agents, chatbots, RAG), giving comprehensive coverage.

5 / 5

Completeness

Clearly states what the skill does and, via a four-part "Use when developers" clause plus a trigger list, explicitly answers when Claude should invoke it with concrete trigger phrases.

5 / 5

Trigger Term Quality

An explicit "Triggers on:" list includes natural developer phrases ("add AI to my app", "build an agent", "tool calling", "useChat") alongside product terms, covering synonyms and API names a user would actually say.

5 / 5

Distinctiveness Conflict Risk

Scoped to a specific product ("AI SDK", "Vercel AI SDK") with distinctive triggers like useChat and ToolLoopAgent, giving it a clear niche with minimal overlap risk against other skills.

5 / 5

Total

20

/

20

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
vercel/vercel-plugin
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.