ax-python-agent-context

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

Quality

78%

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/python/.well-known/agent-skills/ax-python-agent-context/SKILL.md

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.

A concise, well-sectioned overview with a real code sample and sensible guardrails, held back by its failure to deliver the skill's core value: no criteria tell the agent when to choose contextMap versus contextPolicy versus optimization versus recall. The core pattern is also not fully executable since `llm` is never constructed.

Suggestions

Add explicit selection criteria for the four features — e.g., a short decision list mapping task characteristics (persistent corpus vs. within-run compaction, offline tuning vs. runtime recall) to contextMap/contextPolicy/optimization/recall.

Complete the Core Pattern example by constructing the `llm` argument (or show the no-key transport setup) so the snippet is copy-paste runnable.

Convert the Package Facts pointers ("API.md", "axir-api.json", "examples/") into clearly signaled references (e.g., "See [API.md](API.md)") and include the referenced files in the skill bundle so navigation actually resolves.

Dimension	Reasoning	Score
Conciseness	The body is lean and fully sectioned — terse bullets in Package Facts, a four-line core pattern, and guardrails — with no padding and no explanation of concepts Claude already knows; every token earns its place.	5 / 5
Actionability	The Core Pattern snippet uses real package syntax but references an undefined `llm` (not copy-paste runnable), and the skill's central task — "Choose between contextMap, contextPolicy, optimization, and recall" — is given no decision criteria for making the choice, so key execution details are missing beyond anchor 4's "minor gaps".	3 / 5
Workflow Clarity	There is no sequenced procedure or validation checkpoint for the routing decision the skill exists to make; the guardrails supply a few useful conditionals ("if package docs disagree with source code, update the compiler and regenerate packages") but the single action — picking the right feature — remains ambiguous, so the simple-skill exception to reach 5 does not apply.	3 / 5
Progressive Disclosure	Well-organized short sections appropriately split overview from detail, and package resources ("API.md", "axir-api.json", "examples/") are named; however these pointers are listed as package facts rather than clearly signaled navigation links, and no bundle files exist alongside the SKILL.md, leaving minor organization gaps versus anchor 5.	4 / 5
	Total	15 / 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 description: it names a specific domain and enumerates the package's four feature areas with an explicit "Use when" trigger, and its deep package-specific terminology makes it highly distinct. The main gaps are the absence of natural-language synonyms for its jargon and a when-clause that could cite more concrete triggering situations.

Suggestions

Add one or two plain-language synonyms alongside the jargon (e.g., "prompt/program optimization (GEPA/ACE)" or "agent memory") so users who don't know the package's internal terms can still trigger the skill.

Make the when-clause more situational: name the concrete moments the skill applies (e.g., "when building or tuning long-context agents, or choosing how to assemble context for each step").

State the outcome action explicitly (e.g., "selects and wires up the right axllm feature") rather than only "deciding between" feature names.

Dimension	Reasoning	Score
Specificity	Names the domain ("writing Python code with `axllm`") and enumerates several specific capability areas ("context maps, trajectory context policy, offline optimization (ACE/GEPA), and memory recall"), but these are feature nouns rather than explicit actions, so it falls short of the comprehensive anchor 5.	4 / 5
Completeness	Both parts are present: the "what" is deciding between the four named features and the "when" is explicit ("Use when writing Python code with `axllm`..."). The when-clause could name more concrete trigger situations (e.g., building an agent or optimizing a prompt), keeping it at anchor 4 rather than 5.	4 / 5
Trigger Term Quality	Includes multiple terms the target user would naturally say — "Python", "axllm", "ACE/GEPA", "context maps", "memory recall", "long-context" — but provides no synonyms or variations (e.g., plain-language equivalents like prompt optimization or agent memory), matching anchor 4 rather than 5.	4 / 5
Distinctiveness Conflict Risk	Tightly bound to the `axllm` package and niche terms like "trajectory context policy", "ACE/GEPA", and "context maps", giving it a clear niche with distinct triggers and minimal conflict risk with other skills.	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/python/.well-known/agent-skills/ax-python-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.