ax-python-ai

Use when writing Python code with `axllm` for provider clients, model selection, OpenAI-compatible calls, Responses, Gemini, Anthropic, routers, and balancers.

Quality

70%

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

Fix and improve this skill with Tessl

tessl review fix ./packages/python/skills/ax-python-ai/SKILL.md

Quality

Content

65%

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

The body is concise and information-dense with an executable core pattern, but it is held back by reliance on external examples for routing/balancing, absent validation checkpoints, and unresolvable file references plus inline content that would benefit from being split out.

Suggestions

Add an inline minimal executable example for the adaptive balancer (store + reducer + two routes) instead of only pointing to examples/adaptive_balancer_no_key, so the main workflows are copy-paste ready.

Include a short validate/retry checkpoint where relevant (e.g. confirm a route key is non-empty/unique before adding it to a shared store) to sequence the routing workflow with a feedback loop.

Either ship the referenced files (API.md, axir-api.json, examples/) in the bundle or move the API-surface and detailed routing reference into a separate file that does exist, so the signaled references actually resolve.

Dimension	Reasoning	Score
Conciseness	The body is lean and well-organized with no basic-concept padding; each bullet carries package-specific guidance (e.g. 'It combines model lists; it does not learn from outcomes'), so every token earns its place.	3 / 3
Actionability	The Core Pattern is a fully executable snippet and the API surface lists concrete names, but the major Routing And Balancing section relies on external example files instead of inline executable code, leaving the broader workflows incomplete.	2 / 3
Workflow Clarity	Decision rules and a loose progression ('Start with examples/adaptive_balancer_no_key ... then use the cataloged ... example') are present, but there is no explicit validation checkpoint or feedback loop and no crisply sequenced multi-step workflow.	2 / 3
Progressive Disclosure	Sections are well organized and references are clearly signaled and one level deep, but the referenced files (API.md, axir-api.json, axir-capabilities.json, examples/) do not exist in the bundle, and the long API-surface and routing content that could live in a separate reference file is inlined.	2 / 3
	Total	9 / 12 Passed

Description

75%

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 has an explicit trigger and a distinct, package-anchored niche, scoring well on completeness and distinctiveness. It is weaker on specificity and trigger-term quality because it lists capability topics rather than concrete actions and leans on technical jargon over natural user phrasing.

Suggestions

Replace the noun/topic list with concrete verbs, e.g. 'create provider clients, select models, route requests across providers, and balance failover' to raise specificity.

Add natural user phrasings alongside jargon, e.g. 'call OpenAI/Gemini/Anthropic from Python' and 'retry or fail over between providers', so users' actual words match the trigger.

Dimension	Reasoning	Score
Specificity	It names the domain and enumerates capability areas ('provider clients, model selection, OpenAI-compatible calls, Responses, Gemini, Anthropic, routers, and balancers'), but these are topic/noun lists rather than concrete actions like 'extract text' or 'fill forms', so it is not comprehensive enough for a 3.	2 / 3
Completeness	It opens with an explicit 'Use when' trigger clause and states the scope ('writing Python code with `axllm` for ...'), clearly answering both what it does and when to use it.	3 / 3
Trigger Term Quality	It includes some natural terms a user would say ('writing Python code', 'OpenAI', 'Gemini', 'Anthropic') but mixes in heavy jargon ('axllm', 'routers', 'balancers', 'Responses') and misses common variations, matching 'some relevant keywords but missing common variations'.	2 / 3
Distinctiveness Conflict Risk	The description is anchored to a named package (`axllm`) in a specific LLM-provider-coding niche, making it clearly distinguishable and unlikely to trigger for the wrong skill.	3 / 3
	Total	10 / 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: packages/python/skills/ax-python-ai/SKILL.md
Commit: fe8baf8

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.