clickhouse-architecture-advisor
clickhouse/agent-skills
Workload-aware ClickHouse architecture decisioning with provenance-labeled recommendations.
What is clickhouse-architecture-advisor?
This skill guides architectural decisions for ClickHouse systems by identifying workload shape, applying decision frameworks, and attaching official documentation. Use it when designing ingestion patterns, selecting data models, or translating best practices into workload-specific system designs.
- Identifies workload shape (observability, SIEM, analytics, IoT, financial, mixed OLAP)
- Applies decision rule files for ingestion strategy, partitioning, joins, and late-arriving events
- Classifies recommendations as official, derived, or field guidance with explicit provenance
- Attaches official ClickHouse documentation links to every recommendation
- Explains tradeoffs and identifies operating bottlenecks in architecture choices
- Structures responses with workload summary, key decisions, and validation steps
How to install clickhouse-architecture-advisor
npx skills add https://github.com/clickhouse/agent-skills --skill clickhouse-architecture-advisorHow to use clickhouse-architecture-advisor
- 1.Describe your workload shape, latency targets, data characteristics, and query patterns
- 2.The skill identifies the relevant decision rule files (ingestion, partitioning, joins, or upserts)
- 3.Review the structured recommendations with provenance labels (official/derived/field)
- 4.Check attached official documentation links for deeper context
- 5.Validate recommendations using provided SQL, metrics, or smoke tests before implementation
Use cases
- Designing real-time ingestion pipelines with appropriate preaggregation and batching strategies
- Planning time-series retention and partitioning schemes for large-scale observability data
- Selecting join and enrichment patterns for dimension lookups in analytics workloads
- Evaluating mutable state handling and late-arriving event strategies for event streams
- Translating ClickHouse best practices into architecture patterns specific to your data shape and latency targets
- ClickHouse architects designing new systems
- Data engineers selecting ingestion and modeling patterns
- Platform teams building observability or analytics infrastructure
- Teams migrating workloads to ClickHouse and needing decision frameworks
clickhouse-architecture-advisor FAQ
This skill adds workload-aware decision frameworks and explicit provenance labels. It complements best-practices by helping you choose between patterns based on your specific workload shape, latency targets, and operational constraints.
Official = directly from ClickHouse docs; Derived = logically follows from documented behavior but not stated verbatim; Field = experience-based heuristic that may be situational. Field guidance always includes a disclaimer and relevant official docs.
Yes. The skill handles observability, SIEM, product analytics, IoT, financial services, and mixed OLAP workloads. Describe your workload shape and it will apply the appropriate decision rules.
The skill explicitly states when confidence is medium or heuristic, and always provides official documentation links so you can verify against the source of truth.
No. Recommendations distinguish between immediate actions and structural redesign. Prioritize based on your operating bottleneck and validate with provided SQL or metrics before production deployment.
Full instructions (SKILL.md)
Source of truth, from clickhouse/agent-skills.
name: clickhouse-architecture-advisor description: 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. license: Apache-2.0 metadata: author: ClickHouse Inc version: "0.1.0"
ClickHouse Architecture Advisor
This skill adds workload-aware architecture decisioning on top of clickhouse-best-practices.
Official docs remain the source of truth. This skill must always prefer official ClickHouse documentation when available.
Required behavior
Before producing recommendations:
- Identify the workload shape
- observability
- security / SIEM
- product analytics
- IoT / telemetry
- market data / financial services
- mixed OLAP with point-lookups
- Read the relevant decision rule files in
rules/ - Use
mappings/doc_links.yamlto attach official documentation - Classify every recommendation as:
officialderivedfield
- Never present field guidance as official guidance
- If a recommendation is uncertain, say so explicitly
Provenance rules
official
Use this when the recommendation is directly backed by official docs.
derived
Use this when the recommendation is not stated verbatim in docs but follows logically from documented ClickHouse behavior.
field
Use this only for experience-based guidance that may be situational.
When using field, include:
- a disclaimer that the advice is heuristic
- a relevant official doc if one partially applies
- the reason the advice depends on workload context
Read these rule files by scenario
Real-time ingestion design
rules/decision-ingestion-strategy.mdrules/decision-real-time-preaggregation.md- Relevant best-practices insert rules
Time-series and retention design
rules/decision-partitioning-timeseries.md- Relevant best-practices schema partition rules
Enrichment and dimension lookups
rules/decision-join-enrichment.md- Relevant best-practices query join rules
Mutable state / late-arriving events
rules/decision-late-arriving-upserts.md- Relevant best-practices mutation avoidance rules
Output format
Structure responses like this:
## Workload Summary
- workload:
- latency target:
- data shape:
- primary query patterns:
- operational constraints:
## Key Decisions
- ...
- ...
## Recommendations
### <Recommendation title>
**What**
...
**Why**
...
**How**
...
**Category**
official | derived | field
**Confidence**
high | medium | heuristic
**Source**
- doc link(s)
**Validation**
- concrete SQL, metric, or smoke test
Architecture-specific guidance
Prefer decision frameworks over generic advice. Good responses should:
- explain tradeoffs
- identify the likely operating bottleneck
- separate immediate actions from structural redesign
- provide target architecture patterns, not just isolated settings
Full reference
See AGENTS.md for the compiled version and examples/ for sample outputs.
Related skills
More from clickhouse/agent-skills and the wider catalog.
clickhouse-best-practices
31 ClickHouse best-practice rules for schema design, query optimization, and data ingestion.
chdb-sql
>-
clickhousectl-local-dev
Use when a user wants to build an application with ClickHouse, set up a local ClickHouse development environment, install ClickHouse, create a local server, create tables, or start developing with ClickHouse. Covers the full flow from zero to a working local ClickHouse setup.
chdb-datastore
>-
clickhousectl-cloud-deploy
Use when a user wants to deploy ClickHouse to the cloud, go to production, use ClickHouse Cloud, host a managed ClickHouse service, or migrate from a local ClickHouse setup to ClickHouse Cloud.
clickhouse-js-node-troubleshooting
>