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

78%

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/rust/skills/ax-rust-refine/SKILL.md

Quality

Content

68%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-organized reference for the generated axllm package that spends every token on package-specific facts. It stops short of executable guidance — the core snippet uses placeholder variables and real usage is outsourced to an examples/ directory not shipped with the skill — and it presents no sequenced workflow or validation checkpoints for the refinement loop it teaches.

Suggestions

Make the Core Pattern executable: add imports and one complete minimal example that constructs reflection_client, options, and evaluator, or inline a short no-key example instead of only naming symbols.

Add an ordered usage sequence with a validation checkpoint (e.g., verify the call runs against the no-key transport before making provider calls) to support the refinement workflow.

Link the referenced artifacts (API.md, axir-api.json, axir-capabilities.json, examples/) as concrete relative paths so navigation to them is unambiguous.

Dimension	Reasoning	Score
Conciseness	The ~45-line body is all lean bullets, package facts, a two-line snippet, and guardrails — every line carries package-specific information Claude would not know (transports, runtime profile, AxIR rule). No concept explanations, no padding; fits the 'lean and efficient, every token earns its place' anchor.	5 / 5
Actionability	The Core Pattern snippet ('let engine = axllm::AxGEPA::new(reflection_client, options)?') is a call shape with undefined variables rather than executable code, and 'Relevant API Surface' only lists symbol names, deferring real usage to an examples/ directory that is not part of the skill. Not 2 because the pattern and guardrails are concrete pointers; not 4 because nothing in the body is copy-paste runnable or complete.	3 / 5
Workflow Clarity	The body is reference-style (facts, pattern, guardrails) with no ordered usage sequence or validation checkpoints; the implied flow (consult examples, choose transport by credentials, apply the core pattern) and the AxIR-regeneration recovery rule are only implicit, and there is no step to verify generated code actually runs (e.g., against the no-key transport). Not 2 because the guidance that exists is well-defined with guardrail decision rules, not 'many gaps, poorly defined'.	3 / 5
Progressive Disclosure	Well-organized sections for a sub-50-line skill, and detailed materials (API.md, axir-api.json, axir-capabilities.json, examples/) are signaled at one level deep in Package Facts. Not 5 because those references are bare names with no paths or links, and no bundle files accompany the skill, leaving navigation slightly ambiguous; not 3 because nothing that belongs in a separate file is inlined and the structure is clean.	4 / 5
	Total	15 / 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 explicitly names the Rust/axllm niche, lists several concrete capabilities, and includes a clear 'Use when' trigger clause. The main gaps are a few missing trigger synonyms and a what-statement that is embedded in the when-clause rather than stated separately.

Suggestions

Add natural trigger variations users would say ('optimize outputs', 'improve candidates', 'GEPA', 'playbook') to broaden trigger_term_quality coverage.

State the what explicitly before the when clause (e.g., 'Provides Rust API patterns and examples for the axllm package: reward-scored generation, ...') so the capability statement stands on its own.

Dimension	Reasoning	Score
Specificity	The description names the domain ('writing Rust code with `axllm`') and lists several concrete capabilities ('reward-scored generation, iterative candidate improvement, evaluator feedback, and optimizer-backed refinement patterns'). Not 5 because these read as pattern labels rather than comprehensively unpacked actions with slightly jargon-tinged phrasing; not 3 because well more than 1-2 specific actions are named.	4 / 5
Completeness	An explicit 'Use when writing Rust code with `axllm` for...' clause states the trigger, and the capability list conveys the what. Not 5 because the what is folded into the when-clause rather than stated as its own capability sentence, and the trigger is a single formulation rather than the multiple concrete trigger phrases in the 5 anchor; not 3 because both what and when are present and explicit.	4 / 5
Trigger Term Quality	Natural terms a user needing this skill would say are present: 'Rust', 'axllm', 'refinement', 'evaluator feedback', 'optimizer'. A few natural terms/variations are missing (e.g., 'optimize outputs', 'improve candidates', 'GEPA', 'playbook', 'few-shot'), which fits the anchor-4 example of good coverage with a few terms missing rather than the missing-synonyms profile of a 3.	4 / 5
Distinctiveness Conflict Risk	'writing Rust code with `axllm`' pins a specific language and package, and the reward-scored/optimizer refinement capability set is a clear niche with distinct triggers and minimal conflict risk. The anchor-4 case (minor overlap with closely related skills) is not evidenced within the description itself.	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: packages/rust/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.