chdb-datastore

Drop-in pandas replacement with ClickHouse performance. Use `import chdb.datastore as pd` (or `from datastore import DataStore`) and write standard pandas code — same API, 10-100x faster on large datasets. Supports 16+ data sources (MySQL, PostgreSQL, S3, MongoDB, ClickHouse, Iceberg, Delta Lake, etc.) and 10+ file formats (Parquet, CSV, JSON, Arrow, ORC, etc.) with cross-source joins. Use this skill when the user wants to analyze data with pandas-style syntax, speed up slow pandas code, query remote databases or cloud storage as DataFrames, or join data across different sources — even if they don't explicitly mention chdb or DataStore. Do NOT use for raw SQL queries, ClickHouse server administration, or non-Python languages.

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Quality

Content

72%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.

Highly actionable and concise body with strong offloading of bulk API/connector detail to reference files. Weaknesses are the missing validation checkpoint before the batch write and a dead reference to a non-existent examples/examples.md.

Suggestions

Add a validation/debug checkpoint before `.execute()` in the Writing Data example (e.g., call `target.to_sql()` or print the planned operation first), so the batch workflow has an explicit feedback loop.

Create the referenced examples/examples.md file, or remove/fix the broken `examples/examples.md` links in the Decision Tree, Cross-Source Join section, and References list.

Consider adding a short 'verify your environment first' step (run `python scripts/verify_install.py`) at the top of the workflow so the connection examples follow a validated prerequisite.

Dimension	Reasoning	Score
Conciseness	Lean and largely free of concept-padding (no explanations of what pandas/ClickHouse/Parquet are); mostly executable examples and tables. Minor trims possible (e.g., the 10-line API showcase and "Writing Data" block are comprehensive beyond the minimal).	4 / 5
Actionability	Copy-paste-ready code throughout (key insight swap, connection patterns, full pandas API usage, cross-source join, writing data) plus a troubleshooting table with specific commands like `pip install chdb` and `python scripts/verify_install.py`.	5 / 5
Workflow Clarity	The decision tree sequences approaches and troubleshooting offers debug loops (`.to_sql()`, `verify_install.py`), but the "Writing Data" batch `.execute()` operation has no validation checkpoint before execution, capping this dimension at 3 per the batch-operation guideline.	3 / 5
Progressive Disclosure	Good one-level-deep offloading to real files (references/api-reference.md, references/connectors.md, scripts/verify_install.py), but the body and References section repeatedly link to examples/examples.md, which does not exist in the bundle — navigation is partly broken.	3 / 5
	Total	15 / 20 Passed

Description

96%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.

An excellent description: concrete, comprehensive, with explicit what/when triggers and clear negative boundary guidance. The only soft spot is residual overlap with plain pandas skills, which the disambiguation mostly resolves.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ("write standard pandas code", "cross-source joins", "query remote databases or cloud storage as DataFrames", "join data across different sources") plus comprehensive coverage of 16+ sources and 10+ file formats.	5 / 5
Completeness	Explicitly answers both what ("Drop-in pandas replacement with ClickHouse performance... same API, 10-100x faster... cross-source joins") and when ("Use this skill when the user wants to analyze data... speed up slow pandas code... join data across different sources"), plus explicit negative guidance.	5 / 5
Trigger Term Quality	Covers natural user phrasing ("slow pandas code", "pandas-style syntax", "DataFrames", "remote databases", "cloud storage") plus format names (Parquet, CSV, JSON), and even notes triggers apply "even if they don't explicitly mention chdb or DataStore".	5 / 5
Distinctiveness Conflict Risk	Clear niche with explicit "Do NOT use for raw SQL queries, ClickHouse server administration, or non-Python languages" and a named sibling (chdb-sql), but "drop-in pandas replacement" retains minor overlap risk with a generic pandas skill.	4 / 5
	Total	19 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 2 missing	Warning

	Total	15 / 16 Passed

Repository: MapleTechLabs/maple
Path: .agents/skills/chdb-datastore/SKILL.md
Commit: e387359

Reviewed: 1 day 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.