ax-cpp-ai

Use when writing C++ code with `axllm` for named deployment profiles, generic provider clients, model selection, OpenAI-compatible calls, Responses, Gemini, Anthropic, routers, and balancers.

Quality

76%

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-ai/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, information-rich reference that respects the reader's intelligence and gives exact, package-specific rules with almost no fluff. Its weaknesses are the absence of any sequenced workflow for the multi-step tasks it describes, and heavy inlining of detail that belongs in bundled reference files — especially since the paths it cites are not part of the skill bundle itself.

Suggestions

Split deep detail (Typesafe/Jev scoring rules, per-model sampling rules, Astra session work) into reference files inside the skill bundle and keep SKILL.md as a concise overview with clearly signaled one-level-deep links.

Add a short ordered getting-started workflow (create client -> select profile/model -> make call -> handle errors) with explicit checkpoints, so common tasks do not require assembling rules scattered across sections.

Verify cited paths (API.md, axir-api.json, examples/) resolve within the skill bundle, or replace them with references that ship with the skill.

Dimension	Reasoning	Score
Conciseness	Dense, terse bullets with no explanation of concepts Claude already knows; every line carries package-specific rules, exact defaults, and exact error messages. Some very long multi-clause sentences (the sampling and model-catalog bullets) could be tightened, keeping it just below fully lean.	4 / 5
Actionability	Provides concrete API names, exact error classes, exact retry formula ("initialDelayMs * backoffFactor ** attempt ... jitter of 0.75 to 1.25"), defaults, and a runnable core code pattern. Gaps: only one inline code snippet, with most syntax delegated to referenced examples and API docs.	4 / 5
Workflow Clarity	The body is a policy/rules reference with explicit decision rules and failure behavior ("fails before transport", "never retried here", a Guardrails section), but it contains no sequenced multi-step procedure with validation checkpoints for tasks such as setting up Vertex credentials or adaptive balancing. It does not fall to 2 because the guidance that is written is precise and internally consistent.	3 / 5
Progressive Disclosure	Good section headers exist, but large detail blocks (Typesafe scoring rules, Astra session internals, per-model sampling rules) are inlined rather than split into reference files, and the referenced paths (API.md, axir-api.json, examples/, src/examples/cpp/generation/) are not present in this skill bundle, so navigation depends on repo files shipped separately. Structure is present, so it stays above the minimal-structure anchor.	3 / 5
	Total	14 / 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 description: explicit when-to-use triggers, a concrete capability list, and a clearly bounded niche around C++ development with the axllm package. Its main limitation is that capabilities are listed as noun phrases under one shared verb rather than as distinct concrete actions.

Dimension	Reasoning	Score
Specificity	The description lists many concrete capability areas ("named deployment profiles, generic provider clients, model selection, OpenAI-compatible calls, Responses, Gemini, Anthropic, routers, and balancers") that map to the skill body's coverage. It falls short of a 5 because the items are noun phrases sharing the single verb "writing C++ code ... for" rather than multiple distinct concrete actions.	4 / 5
Completeness	The explicit "Use when writing C++ code with `axllm` for ..." clause answers both what and when with concrete trigger phrases. The what is embedded in the same clause as a capability list rather than stated as separate concrete actions, so it does not fully match the anchor requiring both clearly and separately explicit.	4 / 5
Trigger Term Quality	Includes natural terms users would say such as "C++", "OpenAI-compatible", "Gemini", "Anthropic", "routers", and "balancers". A few natural variations are missing (e.g., "Claude", "LLM API", ".cpp"), keeping it below comprehensive synonym coverage.	4 / 5
Distinctiveness Conflict Risk	"C++ code with `axllm`" plus named providers (OpenAI-compatible, Responses, Gemini, Anthropic) pins a clear niche with distinct triggers and minimal risk of firing for an unrelated skill.	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: website/static/cpp/.well-known/agent-skills/ax-cpp-ai/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.