ax-java-gepa

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

Quality

75%

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/java/.well-known/agent-skills/ax-java-gepa/SKILL.md

Quality

Content

76%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 disciplined, token-efficient reference card that assumes Claude's competence and correctly pushes bulk API detail out of SKILL.md. Its main weaknesses are a code snippet that is illustrative rather than executable and the lack of any sequenced optimizer workflow with validation checkpoints.

Suggestions

Add a short numbered workflow for the common case (e.g., 1. pick example from `examples/`, 2. construct reflectionClient, 3. run `engine.optimize`, 4. inspect Pareto front / artifacts before proceeding) with an explicit validation step, which would lift workflow clarity above 3.

Make the Core Pattern self-contained enough to adapt: show how `reflectionClient`, `request`, and `evaluator` are obtained (one import line plus a minimal construction) or explicitly label the snippet as a shape to be filled in from a named example file.

Turn the Package Facts file mentions into clearly signaled links (e.g., 'API docs: see [API.md](API.md); runnable examples: see [examples/](examples/)') and ensure the referenced files actually ship alongside SKILL.md so navigation from the skill works.

Dimension	Reasoning	Score
Conciseness	The body is a lean fact card — intro, When To Use, Package Facts, one two-line snippet, an API surface list, and guardrails — with zero padding and no explanation of concepts Claude already knows. Every section earns its place, matching anchor 5.	5 / 5
Actionability	The Core Pattern is real Java ('AxGEPA engine = new AxGEPA(reflectionClient, java.util.Map.of())' then 'engine.optimize(request, evaluator)') and the API surface names concrete symbols, but the snippet uses undefined variables (reflectionClient, request, evaluator) with no imports, so it is not copy-paste executable; exact syntax is deferred to `examples/`. Concrete code with minor gaps fits anchor 4, not anchor 5's copy-paste-ready bar.	4 / 5
Workflow Clarity	A rough ordering hint exists ('Use BootstrapFewShot before GEPA when demonstrations should seed optimization') and the core call pattern is shown, but there is no sequenced workflow for running the optimizer and no validation checkpoints (e.g., verifying candidate/artifact state or budget limits before proceeding). This is a multi-concern optimizer skill rather than a single-action simple skill, so the simple-skill exception does not apply; anchor 3 (sequence present, checkpoints missing) is the best fit.	3 / 5
Progressive Disclosure	A well-organized single-file overview that points to external detail ('API.md', 'axir-api.json', 'axir-capabilities.json', 'examples/') instead of inlining an API reference, which is good disclosure practice. However, the references appear as bare backtick facts in a 'Package Facts' list rather than clearly signaled navigation, and none of the referenced files ship in this bundle, so it falls short of anchor 5's well-signaled, navigable references.	4 / 5
	Total	16 / 20 Passed

Description

65%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 highly distinct, trigger-rich description for a narrow niche, held back by the absence of an explicit 'what': it tells the agent when to activate but never states what the skill actually does beyond an implied topic list. The topic nouns add specificity of domain but not of action.

Suggestions

Add an explicit 'what' clause naming concrete actions, e.g. 'Runs the GEPA optimizer, configures reflection clients, and tracks metric budgets and Pareto fronts. Use when writing Java code with `dev.axllm:ax`…' — this would raise both specificity and completeness.

Convert one or two topic nouns into verb phrases ('tracks metric budgets and optimizer state' instead of listing 'metric budgets, optimizer state') so the description describes actions, not just subject areas.

Include one or two natural synonyms users might say (e.g., 'prompt optimization' or 'LLM optimizer') alongside 'GEPA' to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	The description names the domain clearly ('writing Java code with `dev.axllm:ax`') but 'writing Java code' is the only concrete action; 'GEPA, Pareto tradeoffs, reflection clients, metric budgets, optimizer state, and artifacts' are topic nouns rather than actions, so coverage of what the skill does is not comprehensive. Fits anchor 3 (domain plus 1-2 concrete actions), not anchor 4 which requires several specific actions.	3 / 5
Completeness	An explicit and specific 'Use when writing Java code with `dev.axllm:ax` for…' trigger is present, but the 'what' is never stated as an action — it is only implied by the enumerated topic list. This sits between anchor 2 (only 'when' without 'what') and anchor 4 (both explicitly stated), landing at the midpoint because the implied 'what' is genuinely informative but not explicit.	3 / 5
Trigger Term Quality	Good natural keyword coverage for the niche — 'Java code', 'GEPA', 'Pareto tradeoffs', 'reflection clients', 'metric budgets', 'optimizer state' — plus the package coordinate '`dev.axllm:ax`'. A few natural terms users might say are missing (e.g., 'prompt optimization', 'LLM evaluation'), so it falls short of anchor 5's synonym-and-extension completeness but is above anchor 3's partial coverage.	4 / 5
Distinctiveness Conflict Risk	'Java code with `dev.axllm:ax` for GEPA' carves out a clear niche with distinct trigger terms; it is unlikely to fire for generic Java, TypeScript, or non-Ax optimization skills. Matches anchor 5 (clear niche, distinct triggers, minimal conflict risk).	5 / 5
	Total	15 / 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/java/.well-known/agent-skills/ax-java-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.