datalad

Retrieve, version, and publish scientific datasets with DataLad and git-annex, and capture computational provenance with datalad run, rerun, and containers-run. Use when cloning or fetching data from OpenNeuro, DANDI, datasets.datalad.org, or any DataLad dataset; when a file in a dataset reads as a broken symlink or a small pointer instead of real data; when an analysis needs a machine-readable record of how each output was produced so it can be re-executed; or when publishing a dataset to siblings such as a GitHub repository plus a storage remote. Also use to decide between DataLad and plain Git for a data-carrying repository.

Quality

100%

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

Quality

Content

100%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 content is high quality: concise and executable, with copy-paste commands, explicit validation checkpoints around destructive operations (drop, run), and clean progressive disclosure to three real reference files. It avoids explaining concepts Claude already knows and instead focuses on the non-obvious git-annex pointer mechanism and DataLad-specific failure modes.

Dimension	Reasoning	Score
Conciseness	Lean and information-dense throughout — it explains only the non-obvious git-annex pointer/content split and provenance mechanism rather than basic Git, and every command example earns its place; not below because there is no padding or over-explanation of concepts Claude already knows.	5 / 5
Actionability	Provides copy-paste-ready, executable commands across all major workflows (clone/get, run with --input/--output, containers-run, publishing via create-sibling-github + initremote) with real flags covering common cases; not below because guidance is concrete and complete rather than pseudocode.	5 / 5
Workflow Clarity	Multi-step processes are clearly sequenced with explicit validation checkpoints — 'datalad wtf --section dependencies' to confirm the environment, '--dry-run' before committing a run, and 'git annex whereis' to confirm another copy exists before 'datalad drop' — and the destructive '--reckless kill' is flagged as a last resort; not below because the symptom/cause/fix table provides explicit feedback loops for error recovery.	5 / 5
Progressive Disclosure	SKILL.md is a clear overview with well-signaled one-level-deep references to the three real reference files (data-access.md, provenance.md, publishing.md), each linked inline and enumerated in a Detailed references section; not below because navigation is easy and content is appropriately split with no nested-reference chains.	5 / 5
	Total	20 / 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.

The description is strong: it states concrete capabilities, provides multiple explicit and natural 'Use when' trigger clauses covering both normal workflows and a distinctive failure symptom, and occupies a clear niche unlikely to conflict with other skills. Both the 'what' and 'when' are answered concretely.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Retrieve, version, and publish scientific datasets with DataLad and git-annex' and 'capture computational provenance with datalad run, rerun, and containers-run' — matching the comprehensive-coverage anchor; not below because coverage spans the tool's main capabilities rather than 1-2 actions.	5 / 5
Completeness	Explicitly answers 'what' (retrieve/version/publish datasets, capture provenance) and 'when' with multiple concrete 'Use when...' trigger clauses plus an 'Also use to decide between DataLad and plain Git' clause; not below because both what and when are explicit with concrete triggers.	5 / 5
Trigger Term Quality	Covers natural user phrases and synonyms including symptom triggers ('broken symlink or a small pointer instead of real data'), verbs ('cloning or fetching', 'publishing'), and named sources (OpenNeuro, DANDI, datasets.datalad.org); not below because it includes both the natural verbs and the concrete failure-mode phrasing a user would actually say.	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (DataLad/git-annex scientific data management) with distinctive triggers like named registries and the broken-symlink symptom; minimal conflict risk with other skills.	5 / 5
	Total	20 / 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: K-Dense-AI/scientific-agent-skills
Path: skills/datalad/SKILL.md
Commit: 49c6e97

Reviewed: 3 days 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.