CtrlK
BlogDocsLog inGet started
Tessl Logo

ax-java-agent-rlm

Use when writing Java code with `dev.axllm:ax` for RLM executor loops, AxCodeRuntime sessions, runtime envelopes, process runtimes, and optional runtime profiles.

56

Quality

71%

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-agent-rlm/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

68%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.

An extremely token-efficient, well-organized overview that correctly treats the generated package as the source of truth. Its weaknesses are incompleteness of the inline code example (undefined `llm`, no imports or setup) and the absence of any explicitly sequenced workflow with validation checkpoints for building and verifying a runtime integration.

Suggestions

Make the Core Pattern self-contained — show how `llm` is obtained (e.g. a no-key transport construction line) and the required imports so the example is copy-paste runnable.

Add a short ordered workflow, e.g. 1. check `examples/` for the target runtime profile, 2. wire the session using the Core Pattern, 3. verify with a `no-key` example, 4. only then switch to a provider transport.

Turn the package-fact pointers into explicit navigable references (e.g. 'See the package's `API.md` and `examples/` shipped alongside this skill') so it is clear where each referenced file lives.

DimensionReasoningScore

Conciseness

The 40-line body is lean with zero padding: every line is a package fact, an API name, or a guardrail Claude could not know on its own. It assumes competence (no explanation of what Java or runtimes are), matching the 'every token earns its place' anchor 5; nothing can be trimmed without losing information.

5 / 5

Actionability

The Core Pattern is real Java rather than pseudocode, but it is not executable: `llm` is undefined, imports and session setup are omitted, and 'Relevant API Surface' lists names without usage. The guardrail 'Start from package examples for exact native syntax' correctly defers detail to the package, but within the body itself key details are missing — anchor 3 ('some concrete guidance but incomplete'), not 4 ('mostly executable with minor gaps').

3 / 5

Workflow Clarity

No explicit step sequence is written, though a rough path is implied: consult package examples, apply the Core Pattern, and verify via the 'no-key examples for deterministic local checks' guardrail. That fits anchor 3 ('sequence present but checkpoints missing or implicit') — a validation approach is mentioned but never sequenced as a checkpoint; anchor 4 would require an explicitly ordered flow.

3 / 5

Progressive Disclosure

The under-50-line body is cleanly sectioned (When To Use, Package Facts, Core Pattern, API Surface, Guardrails) and points to external detail ('API.md', 'axir-api.json', 'axir-capabilities.json', 'examples/'). However, those pointers are stated as bare package facts rather than well-signaled navigable references (no link paths or guidance on where the files live relative to the skill), and none exist as bundle files here, so structure is good but navigation falls short of anchor 5.

4 / 5

Total

15

/

20

Passed

Description

62%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 well-scoped, correctly triggered description for a niche package skill: the 'Use when' clause is explicit and the package coordinate makes it distinctive. Its weaknesses are that it lists domain objects rather than concrete actions and leans on package-internal jargon without natural-language synonyms a user might say.

Suggestions

Convert the topic list into concrete actions, e.g. 'Wire AxCodeRuntime sessions and RLM executor loops, build ProcessCodeRuntime actor-code sessions, and configure javascript-quickjs or python-pyodide runtime profiles.'

Add natural trigger variations beyond jargon, such as 'Java agent', 'Ax runtime', 'dev.axllm:ax', or 'RLM executor' so users phrase the request in everyday terms.

Tighten the 'when' clause so the skill only fires when the user mentions the ax package or its runtime concepts, not for generic Java coding requests.

DimensionReasoningScore

Specificity

The description names the domain precisely ('writing Java code with `dev.axllm:ax`') and lists concrete subjects ('RLM executor loops, AxCodeRuntime sessions, runtime envelopes, process runtimes, and optional runtime profiles'), but these are domain objects rather than actions — the only verb is 'writing'. It matches anchor 3 (domain plus concrete items, not comprehensive) better than 4, which expects several distinct actions ('Extracts text... fills forms, converts pages').

3 / 5

Completeness

Both parts are present: 'what' is writing Java code against the listed runtime constructs, and an explicit 'Use when writing Java code with `dev.axllm:ax`' clause supplies the 'when'. However, what and when are folded into a single clause, so the trigger guidance is explicit but not elaborated with distinct concrete trigger phrases, matching anchor 4 rather than 5.

4 / 5

Trigger Term Quality

'Java code' and the package coordinate `dev.axllm:ax` are natural terms for a user already working with this package, but the remaining terms (AxCodeRuntime, RuntimeEnvelope, RLM executor loops) are package-internal jargon, and no synonyms or variations are offered. Anchor 3 ('Some relevant keywords but missing common variations or synonyms') is the best fit; 4 would require broader natural-phrase coverage.

3 / 5

Distinctiveness Conflict Risk

The generated-package coordinate `dev.axllm:ax` carves out a clear niche with minimal overlap, but the trigger 'writing Java code' would plausibly fire for a user asking to write any Java code without mentioning the package. This is 'mostly distinct; minor overlap risk' (anchor 4) rather than the minimal-conflict anchor 5.

4 / 5

Total

14

/

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
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.