ax-rust-refine

Use when writing Rust code with `axllm` for reward-scored generation, iterative candidate improvement, evaluator feedback, and optimizer-backed refinement patterns.

Quality

80%

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 ./website/static/rust/.well-known/agent-skills/ax-rust-refine/SKILL.md

Quality

Content

75%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 tight, well-structured reference skill: it opens with a near-executable core pattern, states verifiable package facts, and closes with genuinely non-obvious guardrails. Main gaps are the placeholder-laden code snippet, no guidance for choosing among the three listed optimizers, and referenced artifacts whose existence and location are unverified.

Suggestions

Replace the placeholder variables in the Core Pattern snippet (reflection_client, options, request, evaluator) with a minimal complete example — e.g., how to construct a reflection client and an OptimizerEvaluator — so the pattern is copy-paste ready.

Add one line to Relevant API Surface differentiating when to use AxPlaybook vs AxBootstrapFewShot vs AxGEPA, so the reader can pick an optimizer without consulting external docs.

Clarify where the referenced artifacts (API.md, axir-api.json, axir-capabilities.json, examples/) live relative to the skill — or ship them as bundle files — so the references are actually navigable.

Dimension	Reasoning	Score
Conciseness	The body is lean and sectioned, with nearly every bullet carrying non-obvious information (package facts, transport options, TypeScript-only API warning). It sits at 4 rather than 5 because the intro paragraph and "When To Use" bullets partially restate the frontmatter description and could be trimmed.	4 / 5
Actionability	Provides a concrete Rust snippet for the core pattern, an exact API symbol list (AxGEPA, optimize, AxPlaybook, etc.), and explicit direction to consult package examples for exact syntax. The snippet's placeholders (reflection_client, options, request, evaluator) leave minor gaps that keep it below the fully copy-paste-ready 5 anchor.	4 / 5
Workflow Clarity	The core action (construct AxGEPA, call optimize) is unambiguous and the guardrails cover key decision points (provider credentials vs no-key, docs-vs-source conflicts). "Relevant API Surface" lists three optimizer options (AxPlaybook, AxBootstrapFewShot, AxGEPA) without guidance on when to choose each, leaving a minor sequencing gap typical of the 4 anchor.	4 / 5
Progressive Disclosure	Well-organized single-file skill under 50 lines with clear section headers, close to the simple-skill exception. It scores 4 rather than 5 because the body cites API.md, axir-api.json, axir-capabilities.json, and examples/ that are not present in the skill's bundle directory, and their location relative to the skill is never stated — a minor navigation gap.	4 / 5
	Total	16 / 20 Passed

Description

78%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: it names a specific niche (Rust + axllm refinement), lists concrete capability areas, and opens with an explicit trigger clause. The only refinement opportunities are separating the "what" from the "when" into distinct statements and adding a few more natural trigger synonyms.

Dimension	Reasoning	Score
Specificity	Names the domain ("Rust code with `axllm`") and several concrete capability areas ("reward-scored generation, iterative candidate improvement, evaluator feedback, and optimizer-backed refinement patterns"), matching the anchor for several specific actions with minor gaps. Not a 5 because the capabilities are noun-phrase domains rather than a comprehensive set of concrete actions, and not a 3 because coverage clearly exceeds 1-2 actions.	4 / 5
Completeness	An explicit "Use when..." clause with concrete trigger contexts is present, and the "what" (writing Rust code with axllm for these refinement patterns) is stated, so both halves are answered. It sits at 4 rather than 5 because what and when are merged into a single sentence instead of a distinct capability statement followed by explicit trigger phrases.	4 / 5
Trigger Term Quality	Good keyword coverage with natural terms a user of this package would say: "Rust", "evaluator feedback", "optimizer", "refinement". A few natural variations are missing (e.g., "improve outputs", "GEPA", "few-shot"), keeping it below the comprehensive-synonyms anchor of 5 but above the 3 anchor's partial coverage.	4 / 5
Distinctiveness Conflict Risk	Clear niche with a named package (`axllm`) and distinct triggers ("reward-scored generation", "optimizer-backed refinement patterns") that virtually no other skill would match, so conflict risk is minimal — a clean fit for the 5 anchor.	5 / 5
	Total	17 / 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
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/rust/.well-known/agent-skills/ax-rust-refine/SKILL.md
Commit: 3f630a0

Reviewed: 3 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.