CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

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, or tools, (2) Want to build AI agents, chatbots, or text generation features, (3) Have questions about AI providers (OpenAI, Anthropic, etc.), streaming, tool calling, or structured output.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

71%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 lean, well-structured lookup skill with executable commands and properly disclosed one-level references. Its main issues are duplicated provider-search instructions, a missing step for detecting the installed SDK version, and grep commands that need '-r' to actually work on directories.

Suggestions

Consolidate the 'Provider-Specific Information' section into the main documentation section — the grep/glob instructions for providers are already stated there — and keep only the genuinely new note about providerOptions.

Add '-r' to the grep commands (e.g., 'grep -r "your query" node_modules/ai/docs/') so they execute successfully against directories, and add a step telling Claude how to check the installed version (e.g., read 'ai' in package.json) before choosing a documentation path.

Reframe the ai@6.0.34 pin as a deprecated/legacy branch ('For older versions...') so the current-version path is the primary instruction and the version number reads as a compatibility note rather than a time-sensitive claim.

DimensionReasoningScore

Conciseness

Mostly efficient with no conceptual padding, but the 'Provider-Specific Information' section repeats grep/glob guidance already covered ('Provider packages... also include bundled docs in their respective node_modules/@ai-sdk/<provider>/docs/ directories'), and the time-sensitive 'ai@6.0.34' version pin is not placed in a deprecated/old-patterns framing.

3 / 5

Actionability

The guidance is concrete and near copy-paste ready ('grep "your query" node_modules/ai/docs/', 'glob "node_modules/ai/docs/**/*.mdx"', 'https://ai-sdk.dev/api/search-docs?q=your_query'), but the grep commands omit '-r' and would fail when run against a directory — a minor executability gap.

4 / 5

Workflow Clarity

The lookup workflow is clearly sequenced with a version-gated branch and numbered steps, but it never tells Claude how to determine which version is installed (e.g., checking package.json or npm ls), leaving a small gap in the branching condition.

4 / 5

Progressive Disclosure

The ~45-line body is well-sectioned with two clearly signaled, one-level-deep references ([Common Errors Reference](references/common-errors.md) and [AI Gateway Reference](references/ai-gateway.md)), both verified to exist and contain the expected content; navigation is easy.

5 / 5

Total

16

/

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: compact, third-person, with an explicit 'Use when' clause enumerating three concrete trigger scenarios and specific API names. Its only weaknesses are mildly generic capability verbs and a few missing natural synonyms such as 'Vercel AI SDK'.

DimensionReasoningScore

Specificity

The description lists several specific, concrete items — 'generateText, streamText, ToolLoopAgent, or tools', 'AI agents, chatbots, or text generation features', 'streaming, tool calling, or structured output' — though the top-level capability verbs ('Answer questions', 'help build AI-powered features') remain somewhat generic, placing it between the 3 and 4 anchors.

4 / 5

Completeness

It explicitly answers both 'what' ('Answer questions about the AI SDK and help build AI-powered features') and 'when' via an explicit 'Use when developers:' clause with three enumerated concrete trigger conditions, matching the anchor-5 example structure.

5 / 5

Trigger Term Quality

Good keyword coverage including exact API names ('generateText', 'streamText'), feature terms ('chatbots', 'tool calling', 'structured output'), and provider names ('OpenAI, Anthropic'), but a few natural synonyms users would say are missing (e.g., 'Vercel AI SDK', 'LLM app', 'useChat').

4 / 5

Distinctiveness Conflict Risk

The 'AI SDK' niche with named functions (generateText, ToolLoopAgent) is highly distinct and unlikely to fire for the wrong skill; the only overlap surface is the broad 'build AI agents, chatbots' phrasing, which is minor.

5 / 5

Total

18

/

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

compozy/compozy

Path

.agents/skills/ai-sdk/SKILL.md

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.