ax-rust-gepa

Use when writing Rust code with `axllm` for GEPA, Pareto tradeoffs, reflection clients, metric budgets, optimizer state, and artifacts.

Quality

74%

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-gepa/SKILL.md

Quality

Content

65%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 well-organized, token-efficient overview whose detail layer is dangling: the code pattern is a placeholder skeleton and every referenced source of specifics (API.md, manifests, examples/) is absent from the bundle. There is also no sequenced workflow with validation checkpoints for optimizer runs.

Suggestions

Replace the placeholder identifiers in Core Pattern with a minimal runnable example that shows how to construct the reflection client, options, request, and evaluator — or bundle the examples/ directory the guardrails tell the agent to start from.

Add a short numbered workflow (check axir-capabilities.json → pick the matching example → adapt to the request → run with the no-key transport for a deterministic check) with an explicit validation checkpoint before using real provider credentials.

Ship the referenced files (API.md, axir-api.json, axir-capabilities.json, examples/) in the bundle, or inline the minimal API details needed to write a first working call.

Dimension	Reasoning	Score
Conciseness	The body is lean throughout: terse fact lists ("Real network support: yes. Scripted no-key transport support: yes."), no explanation of concepts Claude already knows (what GEPA or Rust is), and no padding. Every section earns its tokens.	5 / 5
Actionability	The Core Pattern uses real Rust syntax but with undefined placeholders (`reflection_client`, `options`, `request`, `evaluator`) — a call shape, not executable code — and it defers to "package examples" and `examples/`, which are absent from the bundle. Better than pseudocode but missing key construction details, placing it between anchors 3 and 4 at 3.	3 / 5
Workflow Clarity	No sequenced workflow exists: the only ordering guidance is "Use BootstrapFewShot before GEPA when demonstrations should seed optimization", and the AxIR-regeneration guardrail is an implicit recovery hint rather than an explicit checkpoint. There is no validation or verification step for running optimizers or tracking budgets.	3 / 5
Progressive Disclosure	Sections are well organized and external materials are clearly signaled ("Package API docs: `API.md` and `axir-api.json`", "Capability manifest: `axir-capabilities.json`", "Runnable examples: `examples/`"), but none of those files exist in the bundle. The body is a pure overview whose entire detail layer is missing, so navigation cannot actually reach the referenced depth.	3 / 5
	Total	14 / 20 Passed

Description

73%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 niche, appropriately terse description with an explicit 'Use when' trigger clause and concrete domain keywords. Its main weakness is that the capability statement rests on a single verb ('writing') plus a topic list rather than naming the distinct actions the skill supports.

Dimension	Reasoning	Score
Specificity	Names the domain ("writing Rust code with `axllm`") with one explicit action (writing) plus a specific topic list ("GEPA, Pareto tradeoffs, reflection clients, metric budgets, optimizer state, and artifacts"). Anchor 4 expects several distinct action verbs; only 'writing' is stated, so anchor 3 is the best fit.	3 / 5
Completeness	An explicit "Use when..." clause answers 'when' with concrete triggers, and "writing Rust code with `axllm`" answers 'what'. The 'what' is thin — it never states what the skill does with those topics (run optimizers, track budgets, inspect artifacts) — so it falls between anchors 4 and 5, closer to 4.	4 / 5
Trigger Term Quality	"Rust code", "axllm", "GEPA", "Pareto tradeoffs", "reflection clients", "metric budgets", "optimizer state" are natural terms a user needing this skill would say. Common variations the body itself uses ("optimizer", "playbook", "few-shot", "ax") are missing, so not anchor 5.	4 / 5
Distinctiveness Conflict Risk	"`axllm`", "GEPA", and "Pareto tradeoffs" carve out a clear niche; the description would not trigger for general Rust-coding or generic optimization skills. Minimal conflict risk.	5 / 5
	Total	16 / 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-gepa/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.