debug-clickhouse-queries

Investigate query evaluation failures in the Knowledge Graph synthetic data pipeline. Use when queries fail or return unexpected results after running the evaluate binary.

1.10x

Quality

—

Does it follow best practices?

Impact

84%

1.10x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Quality

Content

92%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a dense, actionable debugging guide that assumes Claude's competence and provides executable SQL and commands plus a clear validated workflow. Its main weakness is monolithic structure with no progressive disclosure via separate reference files.

Suggestions

Extract the long reference sections (e.g., 'Places to investigate' and 'Simulator configuration impact') into a REFERENCES.md file referenced one level deep, keeping SKILL.md as an overview.

Dimension	Reasoning	Score
Conciseness	The body avoids explaining concepts Claude already knows and instead delivers project-specific knowledge (traversal-path trie semantics, edge iteration directions, sampling fallback) plus executable snippets; every section earns its place despite the length.	3 / 3
Actionability	It provides copy-paste-ready SQL diagnostics and concrete `cargo run -p orbit -- query` commands with multiple flag variations, giving fully executable guidance rather than pseudocode.	3 / 3
Workflow Clarity	A 7-step debugging checklist sequences the investigation, hypothesis-testing branches (UNKNOWN_COLUMN / TYPE_MISMATCH / 0 rows) give error-recovery feedback loops, and 'After changes, regenerate' provides a verification checkpoint.	3 / 3
Progressive Disclosure	The skill is a single >50-line file with no bundle reference files for progressive disclosure; sections are well-organized but all content is inline with no one-level-deep references to split material out.	2 / 3
	Total	11 / 12 Passed

Description

75%

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 well-scoped to a distinct niche with an explicit 'Use when' trigger, fully covering the what/when. It could be strengthened by enumerating more concrete actions and using more natural trigger phrasing.

Suggestions

List multiple concrete actions (e.g., 'inspect generated SQL, check sampling metadata, distinguish data vs query bugs') to raise specificity.

Add more natural user-facing trigger phrasing such as 'empty query results' or 'evaluate report shows failures' alongside the current technical terms.

Dimension	Reasoning	Score
Specificity	It names the domain and a core action ("Investigate query evaluation failures in the Knowledge Graph synthetic data pipeline") but lists only a single action verb rather than multiple concrete actions, so it does not reach the comprehensive level.	2 / 3
Completeness	It explicitly states what it does ("Investigate query evaluation failures...") and when to use it ("Use when queries fail or return unexpected results after running the evaluate binary."), answering both what and when.	3 / 3
Trigger Term Quality	Triggers like "queries fail", "unexpected results", and "evaluate binary" are relevant but lean toward technical/internal phrasing and lack common natural variations a user might say.	2 / 3
Distinctiveness Conflict Risk	The niche is highly specific (Knowledge Graph synthetic data pipeline, evaluate binary), making it clearly distinguishable and unlikely to trigger for unrelated skills.	3 / 3
	Total	10 / 12 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: gitlabhq/orbit-knowledge-graph
Commit: 00ca361

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.