ai-sdk-development

TRIGGER when working with ai-sdk which is Laravel official first-party AI SDK. Activate when building, editing AI agents, chatbots, text generation, image generation, audio/TTS, transcription/STT, embeddings, RAG, vector stores, reranking, structured output, streaming, conversation memory, tools, queueing, broadcasting, and provider failover across OpenAI, Anthropic, Gemini, Azure, Groq, xAI, DeepSeek, Mistral, Ollama, ElevenLabs, Cohere, Jina, and VoyageAI. Invoke when the user references ai-sdk, the `Laravel\Ai\` namespace, or this project's AI features — not for Prism PHP or other AI packages used directly.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

76%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 highly actionable, code-rich skill body that gives executable guidance across every capability with a useful decision workflow. It is somewhat verbose (recapped provider lists and patterns) and monolithic in structure, inlining material that would benefit from separate reference files, and it lacks explicit validation checkpoints.

Suggestions

Move the Provider Support table and full per-capability API examples into separate reference files (e.g. PROVIDERS.md, EXAMPLES.md) and link to them from a concise overview to improve progressive disclosure.

Trim redundant recap sections (Key Patterns re-lists namespaces, classes, and commands already shown in examples) to reduce token overhead and raise conciseness.

Add an explicit validation checkpoint in the decision workflow (e.g. 'check the provider support table before calling a capability; an unsupported call throws LogicException') to strengthen workflow clarity for risky operations.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with code examples earning their place, but the long provider-support table, the repeated 'Key Patterns' recap, and re-listed capabilities (Decision Workflow, Key Patterns, Provider Support) restate content Claude could derive, pushing it toward 'mostly efficient but could be tightened'.	3 / 5
Actionability	Nearly every section ships copy-paste-ready, executable PHP covering the common cases (agents, conversation context, structured output, images, audio, transcription, embeddings, reranking, files/stores, attributes, tools, failover, faking), satisfying the 'fully executable; specific examples cover common cases' anchor.	5 / 5
Workflow Clarity	A clear 'Decision Workflow' maps intents to entry points and the docs-search section gives an ordered procedure, but there are no explicit validation/verification checkpoints for risky operations (e.g. unsupported capability throwing LogicException), so it falls just short of the feedback-loop anchor.	4 / 5
Progressive Disclosure	Sections are well-organized and a doc-search mechanism is referenced, but the SKILL.md is monolithic — large API reference tables and example bulk that could live in separate reference files are inlined, and there are no one-level-deep file links, matching 'some structure but could be better organized'.	3 / 5
	Total	15 / 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 highly specific and complete description with explicit what/when guidance and a useful negative boundary, distinguishing it from Prism PHP. Trigger phrasing is somewhat instructional and could be more aligned with natural user speech, and the broad scope introduces minor overlap risk.

Dimension	Reasoning	Score
Specificity	Enumerates many concrete capabilities (text generation, image generation, audio/TTS, transcription/STT, embeddings, RAG, vector stores, reranking, structured output, streaming, conversation memory, tools, queueing, broadcasting, provider failover) plus a named provider list, matching the 'comprehensive coverage' anchor.	5 / 5
Completeness	Explicitly answers 'what' (unified AI SDK across many providers/capabilities) and 'when' (three explicit trigger clauses: 'TRIGGER when working with ai-sdk', 'Activate when building/editing...', 'Invoke when the user references...'), plus a negative boundary ('not for Prism PHP').	5 / 5
Trigger Term Quality	Includes natural user terms like 'ai-sdk', 'AI agents', 'chatbots', 'image generation', 'embeddings', 'RAG' and the `Laravel\Ai\` namespace, but the trigger phrasing ('TRIGGER when', 'Activate when', 'Invoke when') is instructional rather than the natural phrases a user would say; good coverage with a few synonyms missing.	4 / 5
Distinctiveness Conflict Risk	The negative boundary clause ('not for Prism PHP or other AI packages') and specific `Laravel\Ai\` namespace clearly carve a distinct niche, but the very broad capability list keeps overlap risk slightly above minimal.	4 / 5
	Total	18 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: yaojingang/GEOFlow
Path: .agents/skills/ai-sdk-development/SKILL.md
Commit: 9d70db0

Reviewed: 7 days 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.