clickhouse-architecture-advisor

MUST USE when designing ClickHouse architectures, selecting between ingestion or modeling patterns, or translating best practices into workload-specific system designs. Complements clickhouse-best-practices with decision frameworks and explicit provenance labels.

Quality

61%

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 ./.agents/skills/clickhouse-architecture-advisor/SKILL.md

Quality

Content

52%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 well-organized and concise with a sensible workflow and a useful output template, but as delivered the skill is a hollow shell: all substantive decision content lives in bundle files that are absent. Actionability and progressive disclosure are the casualties, since every pointer is a dead end.

Suggestions

Ship the referenced bundle files (rules/decision-*.md, mappings/doc_links.yaml, AGENTS.md, examples/) alongside SKILL.md, or remove the references — as delivered, every path in the body is broken.

Inline at least one worked example (workload shape → key decisions → recommendation with official/derived/field label and a concrete SQL validation) so the skill remains actionable even when the rule files are unavailable.

Replace vague pointers like 'Relevant best-practices insert rules' with the specific rule file names or IDs from clickhouse-best-practices, and show a sample doc_links.yaml entry so step 3 is executable.

Dimension	Reasoning	Score
Conciseness	The body is lean with no padding and no explanations of concepts Claude already knows; the workload list, provenance rules, and output template each earn their tokens. Not 5 because a few lines ('Relevant best-practices insert rules') are vague filler that could be tightened or made specific.	4 / 5
Actionability	Beyond the output-format template there is no executable guidance — no SQL, commands, or worked example — and the core decision logic is deferred to 'rules/decision-*.md' and 'mappings/doc_links.yaml' files that are not present in the bundle. Matches anchor 2 (high-level hints, missing the specific steps to execute); not 3 because the pseudocode-equivalent template alone doesn't supply concrete executable guidance.	2 / 5
Workflow Clarity	A clear numbered sequence (identify workload → read rule files → attach docs → classify provenance → flag uncertainty) plus a scenario-to-file index, with a 'Validation' field acting as an output checkpoint. Not 5 because validation of the skill's own intermediate steps is left implicit and depends on the missing rule files.	4 / 5
Progressive Disclosure	The one-level-deep reference design is right in principle, but scored against the actual bundle every referenced path is broken: 'rules/', 'mappings/doc_links.yaml', 'AGENTS.md', and 'examples/' do not exist, and no references/, scripts/, or assets/ directories ship with the skill. Matches anchor 2; not 3 because the navigation cannot actually lead anywhere as delivered.	2 / 5
	Total	12 / 20 Passed

Description

70%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 well-scoped description with an explicit 'MUST USE when' trigger and clear differentiation from its sibling skill. Its main weakness is that the capability statements stay at the architectural level ('decision frameworks', 'translating best practices') rather than concrete actions.

Dimension	Reasoning	Score
Specificity	Names the domain and one concrete action ('selecting between ingestion or modeling patterns'), but the remaining verbs ('translating best practices', 'decision frameworks') are abstract rather than concrete capabilities. Not 4 because the action list lacks the verb-level specificity of the anchor-4 example; not 2 because it goes well beyond naming the domain.	3 / 5
Completeness	Both are present: an explicit trigger ('MUST USE when designing ClickHouse architectures, selecting between ingestion or modeling patterns...') and a 'what' ('Complements clickhouse-best-practices with decision frameworks and explicit provenance labels'). Not 5 because the 'what' is less concrete than the anchor-5 example's specific capability list.	4 / 5
Trigger Term Quality	Includes natural phrases users would say ('designing ClickHouse architectures', 'ingestion', 'modeling patterns', 'workload'), giving good keyword coverage within the niche. Not 5 because common synonyms such as 'schema design', 'partitioning', or 'replication' are missing.	4 / 5
Distinctiveness Conflict Risk	'Complements clickhouse-best-practices' carves out a distinct niche (architecture decisioning vs. general practices), but the description itself acknowledges the sibling skill and generic ClickHouse design questions could trigger both. Not 5 due to that acknowledged overlap; not 3 because the triggers are architecture-specific.	4 / 5
	Total	15 / 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: MapleTechLabs/maple
Path: .agents/skills/clickhouse-architecture-advisor/SKILL.md
Commit: 1a03fa7

Reviewed: 11 hours 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.