Tessl • The management layer for AI agent skills

SkillPublic

roboflow-data-management

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

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

88%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.

An exemplary reference-style skill: concrete, executable, with explicit validation and error-recovery loops for batch upload operations. The only real costs are repetition across the upload sections and a monolithic structure that inlines catalog material a references/ file could hold.

Suggestions

Consolidate the upload-method guidance so the preference rule and per-call limits (100 files / 1 GB, 2 GB ZIP) are stated once, with the Video Upload and MCP sections linking back rather than restating them.

Move the exhaustive RoboQL filter table and the preprocessing/augmentation catalogs into a references/ file, keeping SKILL.md as an overview with well-signaled one-level-deep pointers.

Dimension	Reasoning	Score
Conciseness	Mostly lean, fact-dense tables with no padding over concepts Claude already knows, but the upload preference rules and limits ('prefer image_upload_files', 100 files / 1 GB, 2 GB ZIP) are restated across the Methods table, Video Upload, and MCP sections and could be consolidated.	4 / 5
Actionability	Fully executable guidance throughout: copy-paste CLI and Python commands, exact MCP tool names, exact terminal statuses, expected error strings ('Image was already annotated'), and concrete filter syntax examples covering the common cases.	5 / 5
Workflow Clarity	Multi-step flows have explicit validation checkpoints (poll image_upload_status until completed or failed, inspect uploadErrors/annotationErrors/warnings, poll every videoId) and genuine feedback loops (retrying a failed chunk by re-uploading the same zip), plus pre-flight checks before destructive or batch operations.	5 / 5
Progressive Disclosure	Well-sectioned body with clearly signaled one-level references to the labeling and cloud-storage skills, but the exhaustive filter table and preprocessing/augmentation catalogs are inlined in a single ~250-line SKILL.md with no bundle files, where reference files could carry the bulk.	4 / 5
	Total	18 / 20 Passed

Description

78%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, highly specific description with excellent distinctiveness and good trigger coverage. Its main weakness is that the capability statement ('what') is folded into the 'Use when' trigger list rather than stated explicitly.

Suggestions

Open with a brief capability statement before the trigger clause, e.g. 'Upload, label, organize, and version datasets on Roboflow. Use when...' so the 'what' is explicit as well as the 'when'.

Add a few natural synonyms to the trigger list such as 'annotate', 'export', or 'train/valid/test splits' to broaden the phrases users would actually say.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('uploading images or videos, labeling, organizing datasets, creating Roboflow projects (detection/segmentation/keypoint/classification)... RoboQL search'), but 'tags, splits, versions' are bare nouns rather than stated actions, leaving minor gaps in coverage.	4 / 5
Completeness	An explicit 'Use when...' clause with concrete triggers is present, but the 'what' is only implicit inside the trigger list rather than stated as a separate capability statement.	4 / 5
Trigger Term Quality	Good coverage of natural phrases users would say ('uploading images or videos', 'labeling', 'tags', 'splits', 'versions'), but misses common synonyms and variations such as 'annotate', 'export', or 'train/valid/test'.	4 / 5
Distinctiveness Conflict Risk	Explicitly names Roboflow, its project types, and RoboQL, establishing a clear niche with distinct triggers and minimal conflict risk with any other skill.	5 / 5
	Total	17 / 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: chatgpt/skills/roboflow-data-management/SKILL.md
Revision: ad4f5c6

Reviewed: 10 hours ago

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.

roboflow-data-management

Table of Contents

Is this your skill?