ax-java-gen

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

Quality

64%

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/java/skills/ax-java-gen/SKILL.md

Quality

Content

42%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 information-dense with non-obvious, API-specific facts and a coherent section structure, but it functions as an inlined reference manual: far too long for SKILL.md, with almost no executable examples and no explicit validation checkpoints. The Astra Session Work and Provider Forward Options detail belongs in separate reference files that the overview links to.

Suggestions

Move the deep behavioral specifications (Astra Session Work, the streaming/caching/renderAudio/date-parsing/retry detail in Provider Forward Options) into one-level-deep reference files (e.g., SESSIONS.md, FORWARD-OPTIONS.md) and keep a short summary plus pointers in SKILL.md.

Add a few small executable Java snippets alongside the Core Pattern for the highest-frequency tasks (attaching a tool, multi-sampling with a result picker, streamingForward with try-with-resources) so the guidance is copy-paste ready rather than descriptive.

Replace jargon-only phrasing with the natural terms a user would say (e.g., "generating structured Java outputs", "sampling multiple responses and picking the best") so the description triggers on real user wording.

Dimension	Reasoning	Score
Conciseness	The body is ~25KB of dense spec prose — e.g., a 20-line streamingForward paragraph, renderAudio minutiae, and timezone-abbreviation rules ("PST is -08:00 all year; ambiguous ones such as BST, IST and CST are rejected") — that goes well beyond an overview; "as in TypeScript" is repeated dozens of times. This is noticeably more than a few trimmable spots, matching the verbose anchor rather than the mostly-efficient one.	2 / 5
Actionability	Concrete guidance is present via exact option keys and semantics (sampleCount, structuredOutputMode, maxRetries, addStreamingAssert, setSampleCount/setResultPicker), but only one two-line executable snippet (the Core Pattern) exists and the skill defers runnable shapes to package examples ("Start from package examples for exact native syntax"), leaving it between the incomplete and mostly-executable anchors.	3 / 5
Workflow Clarity	A loose sequence exists (When To Use → Package Facts → Core Pattern → examples → Guardrails) with implicit verification guidance ("Use no-key examples for deterministic local checks"), but there are no explicit step sequences or validation checkpoints for the many behaviors described — structure present, checkpoints implicit.	3 / 5
Progressive Disclosure	Section headers are present and the body points to API.md, axir-capabilities.json, and examples/, but those are package facts rather than bundle navigation, and the deep API reference detail (the entire Provider Forward Options and Astra Session Work sections) is inlined rather than split into reference files — structure exists yet content that should be separate is inline.	3 / 5
	Total	11 / 20 Passed

Description

71%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 distinctive, well-triggered description with explicit when-guidance and a broad enumeration of capabilities. Its weaknesses are thin natural-language trigger coverage (jargon-heavy terms) and a what-clause that is only implied by the capability list rather than stated.

Dimension	Reasoning	Score
Specificity	The description enumerates several concrete capability areas ("AxGen programs, forward calls, indexed multi-sampling, result pickers, streaming, tools, assertions, traces, usage, and output parsing"), but the terse noun list (e.g., "usage", "traces") lacks the explicit action phrasing of the top anchor, leaving minor coverage gaps.	4 / 5
Completeness	The when is explicit and concrete ("Use when writing Java code with `dev.axllm:ax` for ..."), and the capability list conveys the what, but only implicitly — there is no standalone statement of what the skill does, so the what could be more explicit as in the anchor-4 example rather than the fully paired anchor-5 form.	4 / 5
Trigger Term Quality	"Use when writing Java code with `dev.axllm:ax`" supplies some natural triggers (Java, the package id, streaming, tools), but phrases like "forward calls", "indexed multi-sampling", and "result pickers" are API-internal jargon rather than terms a user would naturally say, and no synonyms or variations are covered.	3 / 5
Distinctiveness Conflict Risk	The niche is tightly anchored by the package id `dev.axllm:ax` and "AxGen" — distinct triggers unlikely to fire for any other Java or LLM skill, matching the clear-niche anchor.	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/java/skills/ax-java-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.