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

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

—

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./packages/python/skills/ax-python-agent-context/SKILL.md

Quality

Content

76%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 a tight, well-structured reference card with executable core code and clearly signaled references. It lacks an explicit decision workflow with validation checkpoints and the referenced bundle files are absent from the skill directory.

Suggestions

Add a short numbered decision workflow (e.g. 1. match task to contextMap/contextPolicy/optimization/recall, 2. start from matching example, 3. verify against axir-capabilities.json) with a validation checkpoint.

Make the core pattern fully copy-paste ready by showing how `llm` is obtained, or explicitly note it must be supplied by the caller.

Include the referenced bundle files (API.md, axir-api.json, axir-capabilities.json, examples/) so the signaled progressive disclosure actually resolves.

Dimension	Reasoning	Score
Conciseness	Lean and efficient: sections are short, assume Claude's competence (e.g. 'Package Facts', 'Relevant API Surface' as compact lists), and every token earns its place with no padding or over-explanation.	5 / 5
Actionability	Provides an executable core pattern and concrete API surface names, but the example leaves `llm` undefined and lacks copy-paste-ready invocation across common cases.	4 / 5
Workflow Clarity	Content is organized as a reference card rather than a sequenced workflow; the Guardrails give ordering hints but there are no numbered steps or validation checkpoints for the listed decisions.	3 / 5
Progressive Disclosure	Well-organized short overview with clearly signaled references (API.md, axir-api.json, axir-capabilities.json, examples/); the referenced bundle files are not actually present, which slightly limits realized navigation.	4 / 5
	Total	16 / 20 Passed

Description

67%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 is specific and tightly scoped to the axllm package, with an explicit 'Use when' trigger and concrete capability list. Its main weakness is trigger-term quality: the terms are technical jargon rather than natural phrases users would actually say.

Suggestions

Add natural-language trigger phrases users would say (e.g. 'Use when a long-context agent needs to choose what to remember, compact, or optimize') alongside the technical terms.

Broaden trigger term coverage with synonyms for non-experts (e.g. 'context management', 'memory', 'prompt compaction').

Consider adding a brief 'what it produces' clause (e.g. 'generates axllm Python code') to sharpen the 'what' answer.

Dimension	Reasoning	Score
Specificity	Names concrete actions ('deciding between context maps, trajectory context policy, offline optimization (ACE/GEPA), and memory recall') covering several specific capabilities, with minor gaps in coverage rather than comprehensive enumeration.	4 / 5
Completeness	Both 'what' (writing Python with axllm for the listed decisions) and 'when' (explicit 'Use when writing Python code with axllm for...') are present, but the trigger could be more explicit on user intent.	4 / 5
Trigger Term Quality	Keywords are domain-specific jargon ('context maps', 'offline optimization (ACE/GEPA)', 'trajectory context policy') rather than natural phrases a user would say; missing common variations or synonyms.	3 / 5
Distinctiveness Conflict Risk	Tightly scoped to the axllm package and a narrow feature set, making overlap with unrelated skills unlikely; minor risk of overlap with other long-context-agent skills.	4 / 5
	Total	15 / 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/python/skills/ax-python-agent-context/SKILL.md
Commit: 3951b05

Reviewed: 5 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.