ax-rust-agent

Use when writing Rust 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

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 ./packages/rust/skills/ax-rust-agent/SKILL.md

Quality

Content

57%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, accurate-feeling behavioral reference for the axllm package with concrete API names and useful guardrails, but it front-loads too much session/MCP behavioral prose into SKILL.md instead of splitting it into reference files, and offers no sequenced workflow with validation checkpoints. Tightening the repeated cross-language comparisons and moving the session spec to a reference file would lift both conciseness and progressive disclosure.

Suggestions

Move the Astra Session Work behavioral spec (lines 39-64) into a dedicated reference file (e.g. references/session-work.md) and keep a 3-5 line summary with a clearly signaled link, so SKILL.md stays an overview.

Consolidate the repeated 'as in TypeScript' cross-references into a single note stating that Rust mirrors TypeScript semantics, then state only the Rust-specific differences once.

Add a short numbered quick-start sequence (create agent -> attach runtime -> register child agents/tools -> forward/stream -> verify with a no-key example) with an explicit validation step using the no-key transport, giving the reader an end-to-end path with a checkpoint.

Dimension	Reasoning	Score
Conciseness	The body is dense with package-specific behavior Claude cannot know (no general-concept padding), but at ~13 KB of prose it restates session semantics repeatedly ('as in TypeScript' appears 8+ times across the Astra Session Work section, lines 39-64) and could be meaningfully tightened. This is more than minor over-explanation, matching anchor 3 rather than 4.	3 / 5
Actionability	Concrete API names and one runnable snippet are present ('axllm::agent("question:string -> answer:string")?', 'agent.streaming_forward(&mut client, values, options, on_delta)', 'with_tool_module("crm", tools)'), giving mostly executable guidance. However the large Astra Session Work section is behavioral prose with no code, and the Core Pattern snippet omits surrounding setup, so it is not copy-paste ready across common cases (not 5).	4 / 5
Workflow Clarity	The document is a topic-organized reference rather than a sequence; guardrails ('Start from package examples for exact native syntax', 'Use no-key examples for deterministic local checks') give direction but there is no end-to-end agent setup sequence and no validation checkpoints. Steps are implicit rather than incoherent, fitting anchor 3 better than 2.	3 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ are absent), and sections are present, but the ~27-line Astra Session Work behavioral spec is inlined where a one-level-deep reference file belongs, and package-file mentions (API.md, examples/) are not clearly signaled links. Some structure with inline content that should be separate matches anchor 3 rather than 4's 'minor organization gaps'.	3 / 5
	Total	13 / 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, niche-pinned description with an explicit 'Use when' trigger and a comprehensive capability list. Its main weakness is that capabilities are named as feature nouns rather than concrete actions, which slightly blurs the 'what does this skill do' answer and leaves a few natural trigger terms (streaming, flows, child agents) unstated.

Suggestions

Rewrite the capability list with action verbs (e.g. 'Create RLM agents with tools, child agents, and MCP clients; stream runs; harvest failures into a persistent playbook') so the what-side is concrete rather than a feature inventory.

Add the natural trigger terms users would actually say that are currently missing: 'child agents', 'streaming', and 'flows', since these are major features documented in the body.

Replace or gloss the internal jargon 'direct-respond executor skipping' with a user-recognizable phrase such as 'skipping the executor stage for no-tool tasks'.

Dimension	Reasoning	Score
Specificity	The description names the domain ('writing Rust code with `axllm`') and lists ten concrete capability areas ('agents, child delegation, tools, MCP, citations, persistent playbook learning, stage instructions, runtime state, final typed responses, and direct-respond executor skipping'). It stops short of anchor 5 because these are feature nouns under one verb rather than distinct concrete actions, leaving what the skill does with each implicit.	4 / 5
Completeness	It opens with an explicit, specific trigger clause ('Use when writing Rust code with `axllm` for...') answering 'when' clearly, and the 'what' is present via the capability list. Not 5 because the what-side lacks concrete action phrasing, and not 3 because 'when' is explicit rather than weakly implied.	4 / 5
Trigger Term Quality	Natural terms a user of this package would say are present ('Rust code', 'axllm', 'agents', 'tools', 'MCP', 'citations', 'playbook'), but prominent body features like streaming, flows, and child agents (as a phrase) are absent and 'direct-respond executor skipping' is internal jargon. Good coverage with a few natural terms missing fits anchor 4, not the synonym-level completeness of 5.	4 / 5
Distinctiveness Conflict Risk	The description is pinned to a clear niche by the package name ('axllm') and distinctive feature terms ('persistent playbook learning', 'direct-respond executor skipping'), making it unlikely to trigger for a generic Rust or agent skill. Minimal conflict risk matches anchor 5.	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: packages/rust/skills/ax-rust-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.