Name: ax-ai
Rating: 72 (1 reviews)
Author: ax-llm

ax-ai

This skill helps an LLM generate correct AI provider setup and configuration code using @ax-llm/ax. Use when the user asks about ai(), providers, models, routing, adaptive balancing, presets, embeddings, batch audio with ai.transcribe() or ai.speak(), extended thinking, context caching, or mentions OpenAI/Anthropic/Google/Azure/DeepSeek/Mistral/Cohere/Reka/Grok with @ax-llm/ax.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Quality

Content

77%

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

The body is highly actionable with copy-paste-ready code and clear do/don't guardrails, but it is a monolithic ~490-line inline reference with no bundle files, and some internal/algorithmic detail and time-sensitive facts could be trimmed or moved to a deprecated section.

Suggestions

Move the adaptive-balancer internals (EWMA, Normal-Inverse-Gamma posterior, Thompson sampling) to an optional reference file; keep only the config knob guidance Claude needs to generate code.

Split the large reference body into bundle reference files (e.g. references/providers.md, references/routing.md) with clearly signaled one-level-deep links from SKILL.md to improve progressive_disclosure.

Isolate time-sensitive facts (DeepSeek deprecation date 2026-07-24, version 23.0.6) in a dedicated 'Deprecated' or 'Version notes' section rather than inline in DeepSeek Notes.

Dimension	Reasoning	Score
Conciseness	Mostly efficient lean code-plus-rules, but includes implementation internals Claude doesn't need to generate code ('EWMA', 'Normal-Inverse-Gamma posterior', 'Thompson sampling') and inline time-sensitive info (DeepSeek deprecation date 2026-07-24, version 23.0.6) not isolated in a deprecated/old-patterns section, which could be tightened.	2 / 3
Actionability	Provides fully executable, copy-paste-ready TypeScript throughout (ai() factory usage, balancer config, chat, transcribe/speak, embeddings, context caching, Bedrock, Vercel SDK, MCP) with concrete option values and named model enums.	3 / 3
Workflow Clarity	For a single-purpose codegen reference the guidance is unambiguous, with explicit guardrails: 'Critical Rules' and 'Do Not Generate' sections plus per-section rules that tell Claude exactly which patterns to use and which to avoid.	3 / 3
Progressive Disclosure	No bundle files exist; all ~490 lines of API reference live inline in SKILL.md as a monolithic wall that could be split into reference files. External example URLs are clearly signaled, but there is no one-level-deep in-bundle reference structure.	2 / 3
	Total	10 / 12 Passed

Description

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.

The description is third-person, concise, and lists concrete capabilities with an explicit 'Use when...' trigger clause covering feature terms and provider names. It clearly communicates both what the skill does and when to invoke it without padding or over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions: 'generate correct AI provider setup and configuration code', plus routing, adaptive balancing, presets, embeddings, batch audio (ai.transcribe()/ai.speak()), extended thinking, and context caching.	3 / 3
Completeness	Explicitly answers both what ('generate correct AI provider setup and configuration code using @ax-llm/ax') and when via a clear 'Use when...' clause enumerating trigger conditions.	3 / 3
Trigger Term Quality	Strong natural-term coverage: ai(), providers, models, routing, adaptive balancing, presets, embeddings, batch audio, extended thinking, context caching, plus named providers (OpenAI/Anthropic/Google/Azure/DeepSeek/Mistral/Cohere/Reka/Grok) a user would actually mention.	3 / 3
Distinctiveness Conflict Risk	Scoped to the @ax-llm/ax package with provider-specific and feature-specific triggers, making it clearly distinguishable and unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: ax-llm/ax
Path: website/static/typescript/.well-known/agent-skills/ax-ai/SKILL.md
Commit: 7911e27

Reviewed: 5 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.