ax-cpp-gepa

Use when writing C++ code with `axllm` for GEPA, Pareto tradeoffs, reflection clients, metric budgets, optimizer state, and artifacts.

Quality

68%

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-gepa/SKILL.md

Quality

Content

71%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 an exemplary lean overview with a clear structure and sensible guardrails, but it functions as a pointer sheet whose pointers don't resolve: the core code pattern is a non-executable skeleton and every referenced artifact (API docs, capability manifest, examples) is absent from the bundle. Shipping the referenced files or inlining one complete runnable example would make it genuinely actionable.

Suggestions

Include the referenced bundle files (API.md, axir-api.json, axir-capabilities.json, examples/) or remove/replace the dangling references with inline content.

Expand the Core Pattern into one complete, compilable example — includes plus construction of reflection_client, options, request, and evaluator — so the primary code snippet is copy-paste ready.

Add a concrete command or code step for 'Track metric budgets, reflection calls, candidate state, and Pareto fronts' so the tracking guidance is executable rather than descriptive.

Dimension	Reasoning	Score
Conciseness	The body is lean throughout: terse fact bullets ("Language: C++.", "Real network support: yes."), a two-line core pattern, and directive guardrails with zero padding or explanation of concepts Claude already knows. Every token earns its place; there is nothing to trim, so the anchor-4 'minor instances of over-explanation' does not apply.	5 / 5
Actionability	The Core Pattern ('axllm::AxGEPA engine(reflection_client, options); auto result = engine.optimize(request, evaluator);') shows the call shape but is not executable — identifiers are undefined, includes are missing, and the guardrail 'Start from package examples' cannot be followed because the referenced 'examples/' is not present in the bundle. This matches 'concrete guidance but incomplete; missing key details'; not 4 because the gaps are significant rather than minor.	3 / 5
Workflow Clarity	The content is clearly organized (When To Use → Core Pattern → API Surface → Guardrails) with sequencing guidance such as 'Use BootstrapFewShot before GEPA when demonstrations should seed optimization'. No validation checkpoints exist, but the skill involves no destructive or batch operations that require them; not 5 because directives like 'Track metric budgets, reflection calls, candidate state, and Pareto fronts' give no how-to.	4 / 5
Progressive Disclosure	References are clearly signaled and one level deep under Package Facts ('API.md and axir-api.json', 'axir-capabilities.json', 'examples/'), but none of these files exist in the provided bundle — scoring against the actual (empty) bundle structure, every detailed reference dangles. This fits 'some structure but could be better organized / unreliable references'; not 4 because unresolved paths break navigation.	3 / 5
	Total	15 / 20 Passed

Description

51%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.

The description has excellent distinctiveness and good trigger terms for a narrow niche, but it is entirely a 'when' clause with no statement of what the skill does, and its only action verb is generic. Adding one or two concrete actions (e.g., 'Run the GEPA optimizer, inspect Pareto artifacts, track metric budgets') would lift both specificity and completeness.

Suggestions

State what the skill does before the trigger clause, e.g., 'Write and run C++ optimization code with the generated `axllm` package: run the GEPA optimizer, seed demonstrations with BootstrapFewShot, inspect Pareto fronts and artifacts.'

Replace the generic 'writing C++ code' framing with the concrete operations the skill supports (run the optimizer, inspect artifacts, track budgets) to raise specificity.

Add a couple of natural synonym terms (e.g., 'optimizer reflection', 'candidate evaluation') so users' phrasing variations still match.

Dimension	Reasoning	Score
Specificity	The description names the domain thoroughly ("C++", "`axllm`", "GEPA, Pareto tradeoffs, reflection clients, metric budgets") but the only action stated is the generic "writing C++ code" — no concrete capability verbs describe what the skill does. This matches the anchor 'Names the domain but actions are minimal or generic'; it is not 3 because no concrete actions are listed, and not 1 because the domain is richly specified.	2 / 5
Completeness	Only the 'when' is present ("Use when writing C++ code with `axllm` for GEPA...") — the description never states what the skill actually does. This exactly matches anchor 2 ('only when is present without what'); it cannot be 3 because that anchor requires a clear 'what', which is absent.	2 / 5
Trigger Term Quality	Good natural keyword coverage for this niche — "GEPA", "Pareto tradeoffs", "reflection clients", "metric budgets", "optimizer state", "artifacts" are the phrases a user of this package would naturally say. Not 5 because no synonyms or variations of these terms are offered.	4 / 5
Distinctiveness Conflict Risk	"C++", "`axllm`", "GEPA", and "Pareto tradeoffs" define a clear niche with distinct triggers; virtually no other skill would match these terms, so conflict risk is minimal.	5 / 5
	Total	13 / 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-gepa/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.