CtrlK
BlogDocsLog inGet started
Tessl Logo

chdb-datastore

Use when the user has tabular data (pandas DataFrame, parquet, csv, Arrow, json) and wants to filter, group, aggregate, join, or speed up slow pandas. Provides chDB DataStore — same pandas API, ClickHouse engine underneath. Also handles reading from S3, MySQL, PostgreSQL, MongoDB, ClickHouse Cloud, Iceberg, Delta Lake as DataFrames and joining across sources. TRIGGER when: user mentions DataFrame, parquet, csv, "fast pandas", "speed up pandas", or cross-source DataFrame joins; user imports `chdb.datastore` or `from datastore import DataStore`. SKIP this skill for raw SQL syntax (use chdb-sql instead), ClickHouse server administration, or non-Python DataStore API work.

76

Quality

95%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

High

Do not use without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

90%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 tightly written, highly actionable overview with excellent progressive disclosure to real bundle files, derailed only by a dangling references/examples.md and a missing validation step in the batch write workflow. Everything else — token efficiency, executable examples, decision routing, troubleshooting — is near ideal.

Suggestions

Create references/examples.md (or examples/examples.md) with the promised join examples, or remove the three references to it — a missing file cited in the References section breaks navigation.

Add a validation checkpoint after the Writing Data .execute() step (e.g., re-open the target DataStore and verify row counts), since it is a batch write operation.

Make the verify_install.py path relative to the skill root (scripts/verify_install.py) instead of the hardcoded 'python agent/skills/chdb-datastore/scripts/verify_install.py'.

DimensionReasoningScore

Conciseness

The body is lean with zero padding: no explanation of pandas or basic concepts, and the one background line ("lazy, ClickHouse-backed pandas replacement... operations compile to optimized SQL") teaches non-obvious, library-specific behavior Claude cannot already know. Every section and code line earns its place.

5 / 5

Actionability

All guidance is copy-paste executable: the one-line import swap, from_file/from_mysql/from_s3/uri constructors, API one-liners, a full cross-source join pipeline, a write pipeline, and a troubleshooting table with exact fixes. Common cases (filter, select, sort, groupby, join, string/datetime accessors) are all covered with runnable code.

5 / 5

Workflow Clarity

The decision tree clearly sequences approach selection and the troubleshooting table provides error-recovery loops (e.g., "Call ds.to_sql() to see the generated SQL and debug", "Join returns empty result → Check key types match"), which matches anchor 4's clear-sequence-most-checkpoints profile. It is not 5 because the batch Writing Data operation has no post-execute() verification checkpoint, and the verify_install.py path in the troubleshooting table assumes a hardcoded install location.

4 / 5

Progressive Disclosure

The body is a genuine overview with well-signaled, one-level-deep references (api-reference.md and connectors.md both exist and are accurately described), matching anchor 4's good-structure profile. It is not 5 because examples/examples.md is referenced three times ("See examples/examples.md #3-5", "More join examples → examples.md", and the References list) but is missing from the bundle, so a whole promised content area is unreachable.

4 / 5

Total

18

/

20

Passed

Description

100%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 exemplary description: concrete capabilities, explicit use-when triggers with synonyms and file formats, and clear skip guidance that disambiguates it from the sibling chdb-sql skill. Third person voice is maintained throughout.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — "filter, group, aggregate, join, or speed up slow pandas" plus "reading from S3, MySQL, PostgreSQL, MongoDB, ClickHouse Cloud, Iceberg, Delta Lake as DataFrames and joining across sources" — giving comprehensive coverage beyond 1-2 verbs. It fits anchor 5 rather than 4 because the source matrix and join capability leave no significant gaps in the stated scope.

5 / 5

Completeness

It opens with an explicit "Use when the user has tabular data... and wants to filter, group, aggregate, join" clause and adds a concrete TRIGGER list, clearly answering both what (pandas-compatible DataStore over ClickHouse) and when. The SKIP clause strengthens the boundary beyond anchor 5's baseline.

5 / 5

Trigger Term Quality

Triggers include natural user phrases ("fast pandas", "speed up pandas", cross-source DataFrame joins), file formats/extensions (parquet, csv, Arrow, json), and the exact import statements users would write. This matches anchor 5's requirement of natural terms, synonyms, and extensions.

5 / 5

Distinctiveness Conflict Risk

It carves a clear niche (pandas-API analysis over external sources) and explicitly disambiguates: "SKIP this skill for raw SQL syntax (use chdb-sql instead), ClickHouse server administration, or non-Python DataStore API work". Conflict risk with sibling skills is minimal.

5 / 5

Total

20

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

relative_links

Relative link issues: 2 missing

Warning

Total

15

/

16

Passed

Repository
chdb-io/chdb
Reviewed

Table of Contents

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.