CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ax-java-llm

Use when writing Java code with `dev.axllm:ax` for using the generated Ax package, factory functions, package docs, examples, and API reference.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./packages/java/skills/ax-java-llm/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

86%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 exemplary index-style router: token-efficient, fully verifiable references, and concrete routing guidance with no padding. The only improvements are micro-level: make the version-check convention actionable and disambiguate the two signature-related table rows.

DimensionReasoningScore

Conciseness

The body is a lean ~30-line router: a task-to-guide table, two sentences of routing advice, and four terse convention bullets. It assumes Claude's competence, explains no background concepts, and every line (including 'TypeScript call shapes are not interchangeable') carries non-obvious information, matching anchor 5.

5 / 5

Actionability

Concrete, executable direction throughout: a table mapping each task to an exact file to read, and named artifacts ('API.md', 'axir-api.json', 'axir-capabilities.json') for exact signatures. Minor gaps keep it below anchor 5 — e.g. 'Check the installed package version against the guide version' gives no command or location for doing the check.

4 / 5

Workflow Clarity

The workflow (pick a guide from the table, read it, then write code following the conventions) is clear, with cross-area guidance ('For an agent with persistent recall, read the agent and memory guides'). It falls short of anchor 5 because some table rows overlap ambiguously — 'decision signatures and native questions' vs 'signatures, fields, schemas, and validation' — leaving route choice slightly underdetermined; no validation loop is needed since there are no destructive or batch operations.

4 / 5

Progressive Disclosure

The SKILL.md is a pure overview with well-signaled, one-level-deep references: all 16 paths in the table (references/quickstart.md through references/ax-java-refine.md) resolve to real files in the references/ directory, navigation is a single task-keyed table, and no detail content is inlined. This matches anchor 5 exactly.

5 / 5

Total

18

/

20

Passed

Description

66%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.

The description has an explicit, well-targeted trigger and is highly distinctive thanks to the package coordinate, but the 'what' reads as a resource inventory rather than a statement of concrete capabilities, and it omits the natural keywords for most of the skill's coverage areas. It is functional but below the quality of the reference good examples.

Suggestions

Rewrite the 'what' clause as concrete actions, e.g. 'Write Java code against the generated Ax package: provider clients, typesafe signatures, structured generation, agents, and workflow orchestration', instead of the circular 'for using the generated Ax package'.

Add natural trigger keywords for the skill's main subareas (agents, signatures, providers, flows/orchestration, prompt optimization) so users phrasing a task in those terms match this skill.

Lead with the capability list and keep the 'Use when...' clause, matching the structure of the reference good examples (actions first, then triggers).

DimensionReasoningScore

Specificity

The description names the domain ('writing Java code with `dev.axllm:ax`') and a single action, then lists resources ('factory functions, package docs, examples, and API reference') rather than multiple concrete actions. It matches anchor 3 (domain plus 1-2 concrete actions, not comprehensive) rather than 4, which requires several specific actions with only minor gaps.

3 / 5

Completeness

Both elements are explicitly present: an explicit trigger ('Use when writing Java code with `dev.axllm:ax`') and a what ('for using the generated Ax package, factory functions, package docs, examples, and API reference'). The 'what' is a loose resource list rather than action-oriented, so it does not reach anchor 5's fully concrete phrasing, but it is above anchor 3, which requires a missing or merely implied 'when'.

4 / 5

Trigger Term Quality

Relevant keywords are present ('Java', 'dev.axllm:ax', 'Ax package', 'factory functions') but common natural variations users would say are missing — the skill covers agents, signatures, providers, workflows, and optimization, none of which appear. This fits anchor 3 (some relevant keywords but missing common variations), not 4's 'good keyword coverage'.

3 / 5

Distinctiveness Conflict Risk

The package qualifier '`dev.axllm:ax`' pins the skill to a clear niche with distinct triggers and minimal conflict risk with other Java or LLM skills, matching anchor 5 exactly rather than anchor 4's 'minor overlap risk'.

5 / 5

Total

15

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

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-llm/SKILL.md

Reviewed

Table of Contents

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.