CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

dask

Distributed computing for larger-than-RAM pandas/NumPy workflows. Use when you need to scale existing pandas/NumPy code beyond memory or across clusters. Best for parallel file processing, distributed ML, integration with existing pandas code. For out-of-core analytics on single machine use vaex; for in-memory speed use polars.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./backend/cli/skills/data-engineering/dask/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

67%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 well-structured, highly actionable overview with real one-level references and clear decision guides for choosing components and schedulers. The main costs are token weight: per-component reference-content bullet lists and inline Best Practices/pattern sections that duplicate the reference bundle, plus a few pseudo-executable snippets with undefined variables.

Suggestions

Collapse each "Reference Documentation ... which includes:" bullet list into a single one-line pointer (e.g. "Details: references/dataframes.md") to save ~30 lines.

Move the inline Best Practices and Common Workflow Patterns sections into references/best-practices.md, keeping only the top rules in SKILL.md.

Make the scheduler and futures snippets self-contained with defined variables so every example is copy-paste executable.

DimensionReasoningScore

Conciseness

The body is mostly operational (code examples, decision guides, concrete numbers like "~100 MB per chunk"), but each of the five components carries a "Reference Documentation ... which includes:" bullet list (~30 lines of table-of-contents content), and the Overview/Best Practices sections restate material available in references. This matches anchor 3 (mostly efficient, some unnecessary explanation, could be tightened) rather than 4, where only minor instances exist.

3 / 5

Actionability

Most code examples are executable and copy-paste ready (read_csv with glob, map_partitions guidance, scheduler switching), but several snippets reference undefined names (`python_function`, `problematic_computation`, `computation`, `large_dataset`), making them illustrative rather than fully runnable. Fits anchor 4 (mostly executable with minor gaps) rather than 5's fully copy-paste-ready bar.

4 / 5

Workflow Clarity

The "Iterative Development Workflow" is a clearly numbered sequence with checkpoints (test small with synchronous scheduler → validate on a 1000-row sample → scale with distributed and monitor via dashboard), and "Common Issues" provides error-recovery guidance. It is not 5 because no explicit validate-then-proceed gates are given for the batch ETL/output-writing patterns, and not 3 since checkpoints are more than implicit.

4 / 5

Progressive Disclosure

Structure is good: all six references exist, are exactly one level deep, clearly signaled per component, and indexed in a final "Reference Files" section. It falls short of 5 because substantial inline content (Best Practices, Common Workflow Patterns, Integration Considerations) duplicates or belongs in the already-provided references, making SKILL.md longer than a lean overview.

4 / 5

Total

15

/

20

Passed

Description

82%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 strong description with explicit what/when structure, good natural trigger terms, and excellent distinctiveness via explicit vaex/polars disambiguation. The only deductions are second-person voice ("Use when you need...") and application-area phrasing rather than fully concrete action verbs.

Suggestions

Rewrite the trigger clause in third person to avoid the second-person penalty, e.g. "Use when scaling pandas/NumPy code beyond memory or across clusters".

Add concrete action verbs and common synonyms/extensions users would say, such as "processes large CSV/Parquet files", "parallelizes dataframes".

DimensionReasoningScore

Specificity

The description names the domain and several concrete application areas ("scale existing pandas/NumPy code beyond memory or across clusters", "parallel file processing, distributed ML"), which fits anchor 4, but it uses second person ("Use when you need to scale...") which the guidelines penalize by 1 point. It is not anchor 5 because items like "distributed ML" name use cases rather than concrete actions.

3 / 5

Completeness

It explicitly answers both what ("Distributed computing for larger-than-RAM pandas/NumPy workflows") and when ("Use when you need to scale existing pandas/NumPy code beyond memory or across clusters") with concrete trigger phrases, matching the anchor 5 example pattern. Not below 5 since neither half is vague or implicit.

5 / 5

Trigger Term Quality

Good natural keyword coverage ("pandas", "NumPy", "parallel file processing", "distributed", "clusters", "out-of-core"), matching anchor 4's 'good keyword coverage; a few natural terms missing'. Missing common synonyms/extensions users might say, e.g. "dataframe", ".csv", ".parquet", "too big for memory".

4 / 5

Distinctiveness Conflict Risk

It carves a clear niche (larger-than-RAM pandas/NumPy scaling) and explicitly disambiguates against neighboring tools ("For out-of-core analytics on single machine use vaex; for in-memory speed use polars"), giving minimal conflict risk. This is a stronger fit than anchor 4, which only tolerates minor overlap.

5 / 5

Total

17

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository

synthetic-sciences/openscience

Path

backend/cli/skills/data-engineering/dask/SKILL.md

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.