cekura-metric-improvement

Use when the user asks to "improve a metric", "run labs", "leave feedback on a metric", "add to labs", "fix metric accuracy", "review metric results", "find misaligned metrics", or "iterate on metric quality". Covers the metric improvement cycle, the feedback workflow, and the labs pipeline used to refine metric accuracy over time.

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./cekura/skills/cekura-metric-improvement/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.

A well-sequenced, mostly actionable skill body with clear workflow steps, a cost guard, and validation/feedback loops. The main weakness is progressive disclosure: a sizeable API reference and several JSON payloads are inlined in SKILL.md rather than moved into a dedicated reference file, despite a references/ directory existing.

Suggestions

Move the 'API Endpoints Reference' table and the process_feedbacks / create_from_call_log JSON payloads into a references/api-reference.md file and link to it from SKILL.md, keeping only the essential endpoints inline.

Render the Step 4-5 validation as an explicit 'validate → review errors → fix → re-run → only proceed when passing' checklist so workflow_clarity reaches the top anchor.

Trim explanatory asides ('This avoids wasting labs iterations...', 'This gives the benefit of...') and the inline ack-tag preamble to tighten conciseness.

Dimension	Reasoning	Score
Conciseness	The body is largely efficient with tight sections, action-oriented lists, and an endpoint table, but includes a few explanatory asides ('This avoids wasting labs iterations...', 'This gives the benefit of...') and an inline ack-tag preamble that could be trimmed.	4 / 5
Actionability	It provides concrete, executable guidance — real endpoints, JSON payloads, parameter names (page_size, agent_id, timestamp__gte), and a pointer to feedback-examples.md — but the 'Step' sections mostly describe what to do rather than giving ready-to-run tool calls for every case.	4 / 5
Workflow Clarity	The labs cycle is clearly sequenced (Steps 1-6) with a dedicated cost guard and validation checkpoints in Steps 4-5, plus an explicit feedback loop ('If validation fails, leave additional feedback and iterate'); it falls short of 5 because some checkpoints are described rather than rendered as an explicit validate→fix→retry checklist.	4 / 5
Progressive Disclosure	There is a 'Reference Files' pointer to references/feedback-examples.md (a real one-level-deep file), but the large inline API Endpoints Reference table and JSON payloads are content that arguably belongs in a separate reference file, so structure is present but not fully split out.	3 / 5
	Total	15 / 20 Passed

Description

83%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 with an explicit 'Use when...' clause and concrete trigger phrases covering what the skill does and when to invoke it. The main weakness is that actions are framed as user triggers rather than enumerated capabilities, and a few natural synonyms are missing.

Suggestions

Reframe some trigger phrases as concrete capability statements (e.g., 'Leaves structured feedback on metric results, runs the labs auto-improve pipeline, and validates improved metrics') to lift specificity.

Add a couple of natural synonyms or the platform qualifier (e.g., 'Cekura metric', 'metric tuning', 'metric calibration') to broaden trigger-term coverage and reduce overlap with sibling skills.

Dimension	Reasoning	Score
Specificity	The description names the domain (metric improvement) and several concrete actions across the cycle — 'improve a metric', 'leave feedback on a metric', 'review metric results', 'iterate on metric quality' — but the actions are phrased as triggers rather than enumerated capabilities, leaving minor gaps in coverage versus the score-5 anchor.	4 / 5
Completeness	It explicitly answers both 'when' ('Use when the user asks to...') with concrete trigger phrases and 'what' ('Covers the metric improvement cycle, the feedback workflow, and the labs pipeline used to refine metric accuracy over time').	5 / 5
Trigger Term Quality	It lists many natural phrases a user would say ('improve a metric', 'run labs', 'fix metric accuracy', 'review metric results') with good coverage, but is missing common synonyms and the platform-specific extension/name hints that would push it to 5.	4 / 5
Distinctiveness Conflict Risk	The Cekura-specific metric/labs niche is mostly distinct and unlikely to trigger the wrong skill, though the broad terms 'improve a metric'/'review metric results' could lightly overlap with a sibling metric-design or eval-design skill.	4 / 5
	Total	17 / 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-metric-improvement/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.