ax-java-agent-context

Use when writing Java code with `dev.axllm:ax` for deciding between context maps, trajectory context policy, offline optimization (ACE/GEPA), and memory recall for long-context agents.

Quality

79%

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-agent-context/SKILL.md

Quality

Content

72%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 compact, information-dense body with a clean structure and a concrete core code pattern, held back by the absence of any actual decision guidance for the skill's stated purpose — choosing among contextMap, contextPolicy, optimization, and recall. What exists is well organized, but the central routing question is never answered.

Suggestions

Add a short decision table or if/then list mapping task characteristics to each feature (e.g., when to use contextMap vs. contextPolicy vs. offline optimization vs. recall), since this choice is the skill's stated purpose.

Include one runnable example per feature area (or pointers to specific entries in `examples/`) so the guidance is executable for all four routing targets, not just the basic agent pattern.

Add a validation checkpoint (e.g., how to verify a chosen configuration with the no-key transport) so the routing decision can be confirmed before committing to it.

Dimension	Reasoning	Score
Conciseness	Lean and dense: sections ('Package Facts', 'Core Pattern', 'Relevant API Surface', 'Guardrails') carry only package-specific facts, a single tight Java example, and no explanation of concepts Claude already knows. Every token earns its place.	5 / 5
Actionability	The core pattern is concrete and executable-looking ('Ax.agent(...)', 'helper.forward(llm, ...)'), but the skill's central promise — choosing between contextMap, contextPolicy, optimization, and recall — has no decision criteria, code, or per-feature guidance ('Choose between contextMap, contextPolicy, optimization, and recall for a task' is a bare pointer). Key details are missing, matching anchor 3 rather than the mostly-executable anchor 4.	3 / 5
Workflow Clarity	'When To Use' bullets gesture at routing but give no sequence, decision criteria, or checkpoints for picking among the four features, and there is no error-recovery guidance. This matches anchor 3 (sequence/checkpoints missing or implicit); the simple-skill exception does not apply because the single action — feature selection — is ambiguous as written.	3 / 5
Progressive Disclosure	Under 50 lines with no bundle files needed, cleanly organized sections, and external pointers ('API.md', 'axir-api.json', 'axir-capabilities.json', 'examples/') clearly signaled as package facts. Well-organized sections alone justify the top score for a skill this size.	5 / 5
	Total	16 / 20 Passed

Description

78%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 strong, specific description with an explicit 'Use when' trigger and a clear niche anchored to a named package and its concrete features. Its main weakness is that the 'what' is expressed as a purpose clause rather than explicit action verbs, and it lacks synonym-level trigger coverage.

Suggestions

Lead with an explicit action statement (e.g., 'Choose between context maps, trajectory context policy, offline optimization (ACE/GEPA), and memory recall when writing Java code with `dev.axllm:ax`') so the 'what' is stated as a concrete action rather than a purpose clause.

Add natural trigger synonyms such as 'prompt optimization', 'context strategy', or 'agent memory' to broaden the natural-phrases users might say.

Dimension	Reasoning	Score
Specificity	Names the domain ('Java code with `dev.axllm:ax`') and four concrete decision areas ('context maps, trajectory context policy, offline optimization (ACE/GEPA), and memory recall'), but describes no concrete actions beyond 'deciding between' them. This sits between anchor 3 (1-2 concrete actions) and anchor 5 (comprehensive concrete actions).	4 / 5
Completeness	The 'when' is explicit ('Use when writing Java code with `dev.axllm:ax`') and the 'what' is conveyed via a purpose clause ('for deciding between context maps, trajectory context policy, offline optimization (ACE/GEPA), and memory recall'). Not 5 because the what is phrased as purpose rather than a leading explicit action statement; clearly above 3 since the when is explicit, not weakly implied.	4 / 5
Trigger Term Quality	Good natural keyword coverage — 'Java code', 'memory recall', 'long-context agents', 'optimization (ACE/GEPA)', 'trajectory context policy' — terms users of this package would naturally say. Not 5 because there are no synonyms or variations (e.g., 'prompt optimization', 'context strategy'); above 3 because coverage is solid rather than partial.	4 / 5
Distinctiveness Conflict Risk	Clear niche with distinct triggers: a specific package coordinate ('dev.axllm:ax') plus named features ('ACE/GEPA', 'trajectory context policy', 'memory recall for long-context agents'). Minimal conflict risk with any other skill.	5 / 5
	Total	17 / 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-agent-context/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.