ax-cpp-gen

Use when writing C++ code with `axllm` for AxGen programs, forward calls, indexed multi-sampling, result pickers, streaming, tools, assertions, traces, usage, and output parsing.

Quality

72%

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/cpp/skills/ax-cpp-gen/SKILL.md

Quality

Content

46%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.

The body is information-dense and accurate, with precise API names, signatures, and error messages, but it functions as a full manual crammed into SKILL.md rather than an overview: only one executable code example, and exhaustive edge-case catalogs inlined in place of progressive-disclosure reference files. Organization within the giant prose sections is weak, with unrelated subtopics merged into paragraph walls.

Suggestions

Split the exhaustive behavior catalogs (Astra Session Work, Provider Forward Options, date parsing, caching, audio rendering) into separate reference files under references/ and keep SKILL.md as a concise overview with clearly signaled one-level-deep pointers, following the Package Facts pattern already used for API.md and examples/.

Add a few complete, executable C++ examples for the main flows the skill names (attaching tools, indexed multi-sampling with a result picker, streaming_forward with a handler) instead of the single two-line Core Pattern, so guidance is copy-paste ready without the external examples bundle.

Break the multi-paragraph walls in Astra Session Work and Provider Forward Options into titled subsections (e.g. Sessions, Cancellation, MCP Tools, Child Agents; Structured Output, Retries, Field Processors, Dates) so each topic is navigable without reading the surrounding prose.

Dimension	Reasoning	Score
Conciseness	There is no padding or re-explanation of concepts Claude already knows — every sentence carries package-specific behavior — but the ~126 lines of run-on, edge-case-dense prose (e.g. the 400+ word `parseDates` paragraph and the JSON serialization rules) noticeably exceed SKILL.md's overview role and could be substantially tightened or offloaded. Mostly efficient content delivered in an unnecessarily monolithic volume, between anchors 2 and 3.	3 / 5
Actionability	Concrete specifics abound (exact option spellings, function signatures like `std::function<std::optional<Value>(const std::string& key, const Value* value)>`, and exact error messages), but only one two-line executable snippet (the Core Pattern) exists for a code-writing skill, and actual syntax is deferred to `examples/` that are not part of this bundle. Some concrete guidance but incomplete for copy-paste use, matching anchor 3 rather than the mostly-executable anchor 4.	3 / 5
Workflow Clarity	The skill is reference-style rather than sequential (no destructive/batch operations requiring validation checkpoints), but sections like "Astra Session Work" and "Provider Forward Options" interleave many unrelated subtopics (sessions, MCP cancellation, child agents, regex semantics, JSON key ordering, date parsing, caching, audio rendering) into multi-paragraph walls that make the structure hard to follow. Topics are covered but organizational gaps leave it at anchor 3 rather than the clearly sequenced anchor 4.	3 / 5
Progressive Disclosure	No bundle files (references/, scripts/, assets/) exist and nothing is deferred: exhaustive behavior catalogs that clearly belong in separate reference files (session semantics, the forward-options catalog, date-parsing rules) are fully inlined in a monolithic deep-dive. Section headers exist and package docs are listed under Package Facts, which keeps this above anchor 1, but the bulk of detailed content is inlined rather than split, matching anchor 2 more than anchor 3.	2 / 5
	Total	11 / 20 Passed

Description

87%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: explicitly states both what the skill covers and when to use it, with an explicit 'Use when' clause naming concrete trigger capabilities tied to a specific package and language. Its only weakness is that capabilities are listed as topic nouns rather than concrete verb-phrased actions, and a few natural synonym triggers are absent.

Dimension	Reasoning	Score
Specificity	The description names the domain ("writing C++ code with `axllm`") and enumerates ten concrete capability areas ("AxGen programs, forward calls, indexed multi-sampling, result pickers, streaming, tools, assertions, traces, usage, and output parsing"), but these are topic nouns rather than the verb-phrased actions of the anchor-5 example. It is above anchor 3 (only 1-2 concrete actions) but not fully action-phrased comprehensive coverage.	4 / 5
Completeness	It explicitly answers both: the what ("writing C++ code with `axllm` for AxGen programs, forward calls, indexed multi-sampling...") and the when ("Use when writing C++ code with `axllm`...") with concrete trigger phrases. This matches the anchor-5 pattern of stating capabilities followed by an explicit 'Use when' clause with specific triggers.	5 / 5
Trigger Term Quality	Terms like "C++", "AxGen", "axllm", "forward calls", "streaming", and "result pickers" are exactly what a user of this framework would say, giving good keyword coverage. A few natural variants are missing (e.g. "cpp", "structured generation", "structured output"), keeping it below the comprehensive-with-synonyms anchor 5.	4 / 5
Distinctiveness Conflict Risk	The named package `axllm` plus the language pin ("C++ code") establishes a clear niche with distinct triggers and minimal conflict risk, even against sibling per-language variants of the same framework.	5 / 5
	Total	18 / 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/cpp/skills/ax-cpp-gen/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.