cekura-eval-design

Use when the user asks to "create an evaluator", "create evals", "create a scenario", "write a test scenario", "design a test case", "test my agent", "build eval coverage", "plan a test suite", "create red team tests", "set up test profiles", "configure conditional actions", "write a conditional action evaluator", "build a deterministic test", "design an IVR test", "IVR navigation test", "write a unit test for a voice agent", "build a regression test", "scripted scenario", "scripted voice test", "structured evaluator", "exact flow test", "sequential conditions", "fixed sequence test", or "run evals". Also for debugging how the testing agent speaks — "why did it read the number as a word", "make it spell digits", "wrong language" — via scenario_language, personality, and XML tags. Covers evaluator design, coverage strategy, test profiles, mock-tool data, conditional actions (deterministic / unit test / regression / IVR flows), and workflow / red-team / edge-case best practices.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

81%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 highly actionable with concrete endpoints, payload skeletons, and a checkpointed workflow, and it structures detail well across real reference files. Its main weakness is length and redundancy in the authoring-mode guidance, plus one orphaned reference and a missing examples/ directory.

Suggestions

Collapse the repeated behavioral-vs-conditional-actions write-path rule (currently restated in The Eval Design Workflow, Choosing Authoring Mode, and Auto-Generation) into a single canonical statement that the other sections link back to.

Add a direct link to references/auto-generation.md from the body's Auto-Generation section so the on-demand reference is reachable one level deep from SKILL.md.

Either create the cited examples/ files (examples/csv-eval-creation.md, examples/workflow-eval.md, examples/red-team-eval.md) or remove the Example Files section to avoid dangling references.

Dimension	Reasoning	Score
Conciseness	At ~7,000 words the body assumes Claude's competence and avoids concept padding, but it restates the same mode→write-path rule across the Workflow, Choosing Authoring Mode, and Auto-Generation sections, leaving noticeable redundancy that could be tightened.	3 / 5
Actionability	Provides concrete API endpoints ('POST /test_framework/v1/scenarios/'), copy-paste-ready JSON payload skeletons, exact XML tag syntax with value ranges, and tool-enablement tables — fully executable guidance covering the common cases.	5 / 5
Workflow Clarity	A 10-step sequenced eval-design workflow with an explicit pre-creation checkpoint, a dedicated post-generation fixup step, a referenced validation checklist, and a terminal 'Run and validate' step provides clear checkpoints and a validate→fix→retry feedback loop for the credit-costly batch creation context.	5 / 5
Progressive Disclosure	The body is an overview with eight well-signaled, one-level-deep references (all real files, each annotated in the Reference Files section), but references/auto-generation.md is only reachable from another reference rather than surfaced from the body, and the cited examples/ directory does not exist.	4 / 5
	Total	17 / 20 Passed

Description

95%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 is trigger-rich and clearly answers both what the skill does and when to use it, with concrete Cekura-scoped capabilities and low conflict risk. Minor specificity gaps keep it just below a perfect score on the capability-enumeration dimension.

Dimension	Reasoning	Score
Specificity	Lists several concrete capabilities ('evaluator design, coverage strategy, test profiles, mock-tool data, conditional actions (deterministic / unit test / regression / IVR flows), and workflow / red-team / edge-case best practices') but stays at category level rather than enumerating exact actions, leaving minor coverage gaps.	4 / 5
Completeness	Explicitly opens with 'Use when the user asks to...' (concrete 'when') and closes with 'Covers evaluator design, coverage strategy, test profiles, mock-tool data, conditional actions...' (clear 'what'), answering both with concrete trigger phrases.	5 / 5
Trigger Term Quality	Comprehensive natural trigger phrases quoted verbatim ('create an evaluator', 'create evals', 'create a scenario', 'design a test case', 'test my agent', 'build eval coverage', 'run evals', 'scripted scenario') plus debugging phrasings ('why did it read the number as a word'), covering synonyms and variations users would actually say.	5 / 5
Distinctiveness Conflict Risk	A tightly scoped Cekura-specific niche ('Cekura evaluators', 'conditional actions', 'scenario_language') with distinct triggers and explicit disambiguation from the metric-design skill, giving minimal conflict risk.	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: cekura-ai/cekura-skills
Path: cekura/skills/cekura-eval-design/SKILL.md
Commit: af2ceb6

Reviewed: 4 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.