Name: roboflow-data-management
Rating: 79.43 (1 reviews)
Author: roboflow

roboflow-data-management

Use when uploading images, labeling, organizing datasets, creating Roboflow projects (detection/segmentation/keypoint/classification), tags, splits, versions, or RoboQL search.

1.09x

Quality

80%

Does it follow best practices?

Impact

95%

1.09x

Average score across 2 eval scenarios

Securityby

Low

Low-risk findings worth noting

Quality

Content

68%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 body is a dense, well-structured reference with concrete commands and clear sequencing, but loses points for sparse code examples, UI-only workflows, and the absence of validation checkpoints on the batch version-generation pipeline.

Suggestions

Add explicit validation/verification steps to the version creation and class-change workflows (e.g., verify split percentages and inspect generated version before use) to satisfy the batch-operation feedback-loop requirement.

Include concrete usage examples or parameter signatures for the listed MCP tools rather than only naming them, so API-driven tasks are actionable.

Move the full RoboQL filter reference and augmentation options into a separate referenced file to make SKILL.md a leaner overview and improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	Mostly efficient reference style using tables and minimal prose that assumes competence; not 5 because the opening "For agents — source-of-truth" blockquote and a few prose sentences (e.g., the video paragraph) could be trimmed.	4 / 5
Actionability	Provides concrete CLI and SDK snippets ("roboflow import -w <workspace> -p <project-id> ...", "project.version(1).download(\"yolov8\")") plus specific RoboQL syntax; not 5 because MCP tools are only listed without usage/parameter examples and several workflows are UI-navigation only.	4 / 5
Workflow Clarity	The Version Creation Pipeline is a clear 5-step numbered sequence, but it and other batch/destructive operations (class changes) lack validation checkpoints or feedback loops, capping this at 3 per the rubric's batch-operation rule; not 4 because explicit verification steps are absent.	3 / 5
Progressive Disclosure	Well-organized sections with one-level-deep related-page references ("roboflow://skills/data-management/labeling", "roboflow://skills/cloud-storage/SKILL") and no bundle files to misstructure; not 5 because the body is a fairly monolithic inlined reference (full RoboQL filter list, augmentation table) rather than split into separate files.	4 / 5
	Total	15 / 20 Passed

Description

92%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 specific, complete, and distinctive, explicitly pairing concrete actions with a clear "Use when" trigger tied to the Roboflow domain. It only loses a point on trigger term quality for lacking synonyms and file extensions.

Suggestions

Add common synonyms and phrasings users might say (e.g., "annotating" alongside "labeling", "exporting datasets", "train/val/test splits") to broaden trigger coverage.

Consider including a relevant file extension or artifact term (e.g., dataset exports like COCO/YOLO) to improve natural-keyword match likelihood.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions ("uploading images, labeling, organizing datasets, creating Roboflow projects (detection/segmentation/keypoint/classification), tags, splits, versions, or RoboQL search"), giving comprehensive coverage; not 4 because the action list is broad and specific rather than having minor gaps.	5 / 5
Completeness	Explicitly answers both what and when via the "Use when [actions]" structure with concrete trigger phrases; matches the anchor 5 example pattern; not 4 because the when-clause is explicit and specific, not merely present.	5 / 5
Trigger Term Quality	Good natural-term coverage ("uploading images", "labeling", "datasets", "tags", "splits", "versions", "RoboQL search"); not 5 because synonyms (e.g., annotate) and file extensions are absent and a few common phrasings (e.g., export) are missing.	4 / 5
Distinctiveness Conflict Risk	Tightly scoped to a Roboflow-specific niche (project types, RoboQL) with distinct triggers and minimal overlap with other skills; not 4 because the Roboflow/RoboQL anchors make conflict risk minimal, not merely low.	5 / 5
	Total	19 / 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: roboflow/computer-vision-skills
Path: skills/data-management/SKILL.md
Commit: f3b2cbd

Reviewed: 10 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.