ax-python-agent

Use when writing Python code with `axllm` for agents, child delegation, tools, MCP, citations, persistent playbook learning, stage instructions, runtime state, final typed responses, and direct-respond executor skipping.

Quality

75%

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/SKILL.md

Quality

Content

65%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 dense, expert-level reference body: zero padding, precise API names, and useful guardrails. Its weaknesses are structural — inlined reference material that should live in a separate file, no sequenced end-to-end workflow with validation checkpoints, and cross-language detail that dilutes the Python focus.

Suggestions

Move the Astra Session Work, Flat Function Namespaces, and Streaming sections into a reference file (e.g. references/advanced.md) and keep one-line pointers plus the Core Pattern in SKILL.md.

Add a short ordered quick-start sequence (create agent -> attach runtime -> register children -> forward) with an explicit validation step such as running a no-key example before touching provider credentials.

Trim non-Python parity notes (Java, Go, C++, Rust, TypeScript comparisons) to a single line, or collect them under a compatibility section, to sharpen the Python focus.

Dimension	Reasoning	Score
Conciseness	The body is dense and fluff-free with no explanation of concepts Claude already knows, and precise declarative statements like "Register child agents before running the parent: add_child_agent(namespace, name, child)" earn their tokens. Not a 5 because cross-language parity minutiae ("Java, C++, and Rust WebSocket adapters track activity when frames arrive") and repeated "as in TypeScript" clauses could be trimmed from a Python-focused skill.	4 / 5
Actionability	Concrete API calls, defaults, and option names appear throughout (agent.streaming_forward(client, values, options), fn(name).namespace("crm"), flat_function_namespace, clarification_shape defaults), and the Core Pattern gives a runnable shape with a guardrail to consult examples/ for exact syntax. Not a 5 because only one code block exists, its llm variable is undefined, and MCP, child agents, and streaming have no runnable snippets.	4 / 5
Workflow Clarity	Sections progress coherently from When To Use through Core Pattern to advanced topics, with a Guardrails section supplying some checks (e.g. "forward requires an executable runtime and fails before any model request if none is supplied"). However, there is no sequenced workflow for building an agent end to end and no validation checkpoints for the multi-part tasks covered; the sequence is implicit.	3 / 5
Progressive Disclosure	The file is well-sectioned and external materials are clearly signaled one level deep (API.md, axir-api.json, axir-capabilities.json, examples/, src/examples/python/generation/), but no bundle files exist and roughly half the body (Astra Session Work, Flat Function Namespaces, Streaming An Agent Run) is inlined API-reference material that belongs in a separate reference file.	3 / 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 solid trigger-style description: explicit 'Use when' guidance, a concrete package anchor, and a broad specific capability list. Main gaps are the absence of action verbs and natural synonyms for the more exotic jargon terms.

Suggestions

Rewrite the capability list with verbs (e.g. 'Build RLM agents, delegate to child agents, attach MCP clients, enforce evidence citations') so the what reads as concrete actions rather than a feature inventory.

Add natural synonyms users might actually say, such as 'subagents', 'AxAgent', or 'LLM agent', alongside the existing jargon terms.

State the trigger more specifically, e.g. 'Use when writing or debugging Python code that imports axllm', to sharpen the when clause.

Dimension	Reasoning	Score
Specificity	The description lists ~10 named capability areas anchored to a concrete package ("child delegation, tools, MCP, citations, persistent playbook learning, stage instructions, runtime state, final typed responses, and direct-respond executor skipping"). It stops short of a 5 because the items are noun-phrase feature names rather than concrete actions, and well above a 3 since coverage extends far past 1-2 actions.	4 / 5
Completeness	The explicit clause "Use when writing Python code with `axllm`" answers when, and the capability inventory answers what. Not a 5 because the what is a feature list rather than a direct statement of what the skill does, and the trigger situation is single-fold (writing axllm Python code).	4 / 5
Trigger Term Quality	Natural terms a user would say are present ("Python", "agents", "tools", "MCP", "citations", "runtime state", "child delegation"). Not a 5 because common synonyms like "subagents", "AxAgent", or "LLM" are missing, and phrases like "direct-respond executor skipping" and "stage instructions" are jargon users would not naturally utter.	4 / 5
Distinctiveness Conflict Risk	The package name `axllm` plus the narrow capability list carves a clear niche with minimal conflict risk. Not a 5 because generic high-traffic terms ("agents", "tools", "MCP", "citations") overlap with many LLM/agent skills if the axllm qualifier is missed.	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/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.