ax-rust-gen

Use when writing Rust code with `axllm` for AxGen programs, forward calls, indexed multi-sampling, result pickers, streaming, tools, assertions, traces, usage, and output parsing.

Quality

63%

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-gen/SKILL.md

Quality

Content

32%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.

The body is an encyclopedic behavior specification: dense, accurate-looking detail about the axllm API, but delivered as multi-hundred-word prose paragraphs inlined into SKILL.md instead of split into reference files, with almost no complete executable examples and no sequenced workflow with validation checkpoints. It functions as a dump of package parity notes rather than an actionable skill overview.

Suggestions

Move the long behavioral specifications (forward options, retries, caching, date parsing, field processors) into one-level-deep reference files under references/ (e.g., references/forward-options.md, references/retries.md) and keep SKILL.md as a concise overview with clearly signaled pointers — the inlined spec both bloats token cost and leaves referenced files like API.md and examples/ absent from the bundle.

Add a small number of complete, executable Rust examples (imports, client construction, forward call, error handling) covering the common cases described in 'When To Use', instead of a single two-line core pattern plus prose signatures.

Give the skill an explicit ordered workflow with checkpoints, e.g., 1) check the package examples for the needed call shape, 2) write the program, 3) validate locally with a no-key scripted example, 4) only then run against a real provider — turning the existing Guardrails hints into a verifiable sequence.

Dimension	Reasoning	Score
Conciseness	The body is a wall of extremely long single-paragraph passages (the streaming paragraph alone runs ~400 words) that inlines behavioral minutiae such as JSON key-ordering rules and timezone DST-gap handling, and repeats 'as in TypeScript' roughly twenty times as parity padding. It is noticeably verbose per anchor 2, though not explaining concepts Claude already knows, which keeps it above anchor 1.	2 / 5
Actionability	There is one minimal core snippet ('axllm::ax("question:string -> answer:string")?' / 'program.forward(&llm, inputs, None)?') without imports or client setup, plus many concrete signatures, defaults, and exact error strings ('streaming_forward(&mut client, values, options, on_delta)', 'maxSteps default 25'). Concrete guidance exists but complete executable examples are missing, matching anchor 3.	3 / 5
Workflow Clarity	Only a rough sequence exists (When To Use bullets, a two-line Core Pattern, then encyclopedic detail), with no validation checkpoints or feedback loops and no ordered steps for the multi-part workflows described; the guardrail 'Start from package examples' is a hint, not a verifiable step. This matches anchor 2: rough sequence, many gaps, validation absent.	2 / 5
Progressive Disclosure	Roughly one hundred lines of reference-grade API behavior specification are inlined directly in SKILL.md — content that clearly belongs in separate reference files per the anchor-2 example — and the referenced artifacts ('API.md', 'axir-api.json', 'axir-capabilities.json', 'examples/') do not exist in the skill bundle, so the references are not backed by real files.	2 / 5
	Total	9 / 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 with an explicit 'Use when...' trigger, a specific package niche, and a broad enumeration of capability areas. Its main weakness is that capabilities are listed as noun phrases folded into the trigger clause rather than as distinct, concrete action statements.

Dimension	Reasoning	Score
Specificity	The description names the domain ('Rust code with `axllm`') and lists many capability areas ('forward calls, indexed multi-sampling, result pickers, streaming, tools, assertions, traces, usage, and output parsing'), but these are noun phrases enumerating feature areas rather than concrete verb-actions, matching anchor 4 rather than 5.	4 / 5
Completeness	The 'when' is explicit and concrete ('Use when writing Rust code with `axllm` for...'), and the 'what' is embedded in the same clause (writing Rust code for the listed feature areas) rather than stated as a distinct action, so it has both but is weaker than the anchor-5 pattern of a separate clear what plus trigger phrases.	4 / 5
Trigger Term Quality	Natural phrases are present ('Use when writing Rust code', 'streaming', 'tools', 'output parsing') alongside package-specific terms a user of this package would say ('AxGen', 'multi-sampling'), but common variations such as 'LLM calls' or 'structured output' are missing, placing it at anchor 4.	4 / 5
Distinctiveness Conflict Risk	The unique package identifiers '`axllm`' and 'AxGen' create a clear niche; a user mentioning them unambiguously targets this skill with minimal overlap risk with any other skill.	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-gen/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.