ax-python-agent-memory-skills

Use when writing Python code with `axllm` for agent memory, recall callbacks, dynamic skill discovery, loaded-skill state, and used-skill tracking.

Quality

73%

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-memory-skills/SKILL.md

Quality

Content

61%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 lean, well-sectioned reference-style body that documents lifecycle semantics accurately without padding. Its weaknesses are actionability and workflow clarity: one incomplete code snippet, no executable how-to steps, and external references that are not part of the skill bundle.

Suggestions

Add one complete, runnable snippet per major task (e.g., seeding constructor skills, registering an onUsedSkills observer, exporting/restoring runtime state) with the `llm` binding and imports included, so the core pattern is not the only executable code.

Add validation checkpoints where mistakes are likely: e.g., after `restore_runtime_state(...)` verify loaded skills match expectations, and a check that the correct state shape (bare vs portable snapshot) is used before save/restore, rather than only the 'Do not interchange the two shapes' warning.

Bundle or inline the referenced material: since `API.md`, `axir-capabilities.json`, and the example files are not in the skill bundle, either include them under references/ or inline the minimal relevant excerpts so the pointers do not dead-end.

Dimension	Reasoning	Score
Conciseness	The body is dense and unpadded: bullet lists of API semantics with no explanation of concepts Claude already knows. It stops short of anchor 5 because the `contextMap` bullet carries time-sensitive version migration detail ("Since 25.0.0... a map saved by a 24.x port...") outside any 'old patterns'/'deprecated' section, which the guidelines penalize.	4 / 5
Actionability	There is one minimal snippet (`helper.forward(llm, {...})`) whose `llm` variable is undefined, and the bulk of the body is declarative API semantics rather than executable steps. The guardrail "Start from package examples for exact native syntax" and the cited example paths are concrete, but `API.md`, `examples/`, and `src/examples/python/...` are not part of the skill bundle, so the skill itself leaves key execution details missing — matching anchor 3 rather than 4.	3 / 5
Workflow Clarity	Topics are clearly sectioned (When To Use, Core Pattern, Lifecycle, Examples, Guardrails), but no task is presented as a step sequence, and validation checkpoints are absent throughout (e.g., "Do not interchange the two shapes" warns about state-shape mixing without any verify step). This sits at anchor 3: organized content with no explicit checkpoints; not anchor 2 since structure and coverage are coherent.	3 / 5
Progressive Disclosure	A well-organized single-file overview with clearly signaled external pointers ("Package API docs: `API.md` and `axir-api.json`", "Runnable examples: `examples/`", the website gallery). It does not reach anchor 5 because none of the referenced files ship in the skill bundle (no references/, scripts/, or assets/ exist), so navigation partially dead-ends for the agent.	4 / 5
	Total	14 / 20 Passed

Description

75%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: explicit trigger clause, specific feature list, and a well-scoped package niche. Its main limitation is that the what and when are merged into a single clause, and it misses a few natural trigger synonyms that would broaden recall.

Dimension	Reasoning	Score
Specificity	The description lists several concrete, specific actions ("agent memory, recall callbacks, dynamic skill discovery, loaded-skill state, and used-skill tracking") anchored to a named package. It falls short of anchor 5 because the actions are feature-area names rather than a comprehensive set of concrete operations (e.g., state export/restore, catalog search are covered by the skill but not the description).	4 / 5
Completeness	It has an explicit "Use when..." trigger clause and a discernible what ("writing Python code with `axllm`" for the listed features), satisfying anchor 4. It does not reach anchor 5 because the what is folded into the when-clause rather than stated as a distinct, concrete capability sentence alongside the trigger.	4 / 5
Trigger Term Quality	Good keyword coverage with natural terms users would say ("agent memory", "skill discovery", "tracking") plus the package name `axllm`. A few natural synonyms are missing (e.g., "remember", "persist", "skill catalog"), keeping it below anchor 5's comprehensive synonym coverage.	4 / 5
Distinctiveness Conflict Risk	Tied to a niche package with feature-specific triggers ("recall callbacks", "loaded-skill state", "used-skill tracking"), giving it a clear identity. Minor overlap risk remains with sibling `axllm` Python skills and generic "agent memory" requests, so it does not fully meet anchor 5's minimal-conflict standard.	4 / 5
	Total	16 / 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-memory-skills/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.