ax-cpp-typesafe

Use when writing C++ code with `axllm` for Typesafe Jev boolean/class signatures, value descriptions, configurable Noul conversion, native Noul/Choice/Score, structured criteria and hybrid generation.

Quality

74%

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/cpp/.well-known/agent-skills/ax-cpp-typesafe/SKILL.md

Quality

Content

67%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 well-structured, information-dense reference skill with one strong executable core example and clear routing guidance. Its main weaknesses are the dense run-on spec prose and the large inline API symbol list that duplicates API.md.

Suggestions

Convert the 'Typesafe / Jev' prose into short labeled bullets (Thresholds, Probability constraints, Balancer behavior) so constraints are scannable instead of buried in run-on paragraphs.

Cut the 'Relevant API Surface' list to the handful of symbols used in examples and defer the rest to API.md, which the body already points to.

Add a minimal native-client code snippet or a one-line 'start here' pointer to the examples directory so the native and hybrid workflows are as actionable as the Jev core pattern.

Dimension	Reasoning	Score
Conciseness	No padding with concepts Claude already knows, but the 'Typesafe / Jev' section is seven paragraphs of compressed spec prose and line 57 inlines ~40 API symbols even though the body itself delegates API docs to 'API.md'. Mostly efficient, but could be tightened considerably.	3 / 5
Actionability	The Core Pattern block is complete, executable code covering the main use case, and the guardrails direct the agent to runnable examples for exact syntax. Not 5 because the extensively discussed native client (system_one, list_models) never appears in code and there are no build/run steps.	4 / 5
Workflow Clarity	'When To Use' gives unambiguous routing (Jev for typed decisions, native for probabilities/scoring, separate program for prose) and Guardrails cover fallback and degradation. Not 5 because there is no setup or first-step sequence telling the agent where to begin.	4 / 5
Progressive Disclosure	Sections are well organized and Package Facts clearly signals one-level-deep references (API.md, axir-api.json, axir-capabilities.json, examples/, external URL). The inline 'Relevant API Surface' list works as a compact index with bulk delegated to API.md, though at ~40 symbols it stretches past 'a few key items inline'.	4 / 5
	Total	15 / 20 Passed

Description

71%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 distinctive, trigger-explicit description for a narrow niche, weakened by heavy reliance on invented project jargon. Users of the package would find it, but the capability list reads as a feature inventory rather than natural actions.

Suggestions

Add natural-language synonyms alongside the jargon, e.g. 'typed boolean/class outputs, classification (Choice), scoring rubrics (Score)' so users who don't know the internal terms can still match it.

Lead the what-list with an action verb, e.g. 'Generate typed C++ decision programs with axllm...' instead of a bare feature inventory.

Trim low-value trigger terms like 'configurable Noul conversion' and 'structured criteria' that a user would rarely say verbatim.

Dimension	Reasoning	Score
Specificity	Lists several concrete capability areas ('Typesafe Jev boolean/class signatures', 'value descriptions', 'configurable Noul conversion', 'native Noul/Choice/Score', 'hybrid generation') with minor gaps. Not 5 because these are noun-phrase features rather than full concrete actions like 'extract text and tables'.	4 / 5
Completeness	Both parts are present: an explicit trigger ('Use when writing C++ code with `axllm`') and a what-list of capabilities. Not 5 because the 'what' is a cryptic feature inventory with no action verb, leaving what the skill actually does for you only implicitly clear.	4 / 5
Trigger Term Quality	'C++ code' and 'axllm' are natural terms a user would say, but 'Typesafe Jev', 'Noul', and 'Noul/Choice/Score' are project-internal jargon, and common natural synonyms (typed outputs, classification, scoring) are missing. Some relevant keywords, but coverage of natural phrasing is incomplete.	3 / 5
Distinctiveness Conflict Risk	The combination of C++ with the specific `axllm` package and Typesafe Jev features carves out a clear niche with distinct triggers; it would virtually never fire for an unrelated skill.	5 / 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/cpp/.well-known/agent-skills/ax-cpp-typesafe/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.