Name: ml-pipeline-workflow
Rating: 37.6 (1 reviews)
Author: administrakt0r

Blog Docs Log in Get started

ml-pipeline-workflow

Complete end-to-end MLOps pipeline orchestration from data preparation through model deployment.

Quality

35%

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

Fix and improve this skill with Tessl

tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills/skills/ml-pipeline-workflow/SKILL.md

The canonical home for this skill is ml-pipeline-workflow in administrakt0r/AI-Agents-Safe-Coding-Skills

Quality

Content

35%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill reads more like a high-level MLOps overview document than an actionable skill for Claude. It catalogs well-known concepts, tools, and patterns without providing the concrete, executable guidance needed to actually build a pipeline. The extensive bullet-point lists of things Claude already knows consume significant token budget while the code examples are stubs that defer to non-existent reference files.

Suggestions

Replace the abstract descriptions and tool catalogs with concrete, executable code examples — e.g., a complete minimal Airflow DAG or Dagster pipeline definition that Claude can adapt, rather than a Python list of stage name strings.

Add explicit validation checkpoints with concrete commands/criteria at each pipeline stage (e.g., 'Run `great_expectations checkpoint run data_quality` — only proceed if all expectations pass').

Remove sections that explain concepts Claude already knows (what A/B testing is, what canary deployments are, lists of cloud platforms) and focus on project-specific patterns, gotchas, and non-obvious configuration details.

Either provide the referenced bundle files (references/*.md, assets/*.yaml.template) or inline the essential content — currently the skill defers to files that don't exist, leaving Claude with no actionable detail.

Dimension	Reasoning	Score
Conciseness	The skill is heavily padded with high-level descriptions of concepts Claude already knows (what MLOps is, what A/B testing is, what canary deployments are). Sections like 'What This Skill Provides' and 'Integration Points' are essentially catalogs of well-known tools and concepts that add no actionable value. The content could be reduced by 60%+ without losing useful information.	2 / 5
Actionability	Despite claiming to provide 'actionable steps,' the content is almost entirely abstract descriptions and bullet-point lists. The code examples are either trivial (a Python list of stage names), incomplete stubs with comments like '# See references/...', or pseudocode YAML. There is no executable, copy-paste-ready code for any actual pipeline implementation.	2 / 5
Workflow Clarity	The Production Workflow section provides a clear 4-phase sequence with sub-steps, and the Debugging Steps section offers a reasonable troubleshooting sequence. However, there are no explicit validation checkpoints with feedback loops — the validation phase is described abstractly without concrete commands or criteria for pass/fail decisions. For a pipeline orchestration skill involving potentially destructive batch operations, the lack of concrete validation steps is a significant gap.	3 / 5
Progressive Disclosure	The skill references external files in 'references/' and 'assets/' directories with clear descriptions, which is good structure in principle. However, no bundle files are provided, so these references are unverifiable dead ends. The main file itself is bloated with content that should either be in those referenced files or omitted entirely, undermining the progressive disclosure pattern.	3 / 5
	Total	10 / 20 Passed

Description

36%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 identifies the MLOps domain but remains at a high level of abstraction, listing pipeline phases rather than concrete actions. It completely lacks a 'Use when...' clause, making it difficult for Claude to know when to select this skill over related ones. The description would benefit significantly from specific actions and explicit trigger guidance.

Suggestions

Add a 'Use when...' clause with trigger terms like 'ML pipeline', 'model training', 'model deployment', 'experiment tracking', 'feature store', or 'model registry'.

Replace the abstract 'end-to-end' phrasing with specific concrete actions such as 'Trains and evaluates ML models, versions datasets, manages experiment tracking, deploys models to serving endpoints, and monitors model performance'.

Include natural synonyms and related terms users might say, such as 'machine learning', 'ML workflow', 'model serving', 'CI/CD for ML', '.pkl', '.onnx', or specific framework names.

Dimension	Reasoning	Score
Specificity	Names the domain (MLOps) and gestures at a pipeline ('data preparation through model deployment'), but the actions are abstract phases rather than concrete operations. No specific actions like 'train models', 'version datasets', 'configure CI/CD', etc. are listed.	2 / 5
Completeness	Provides a vague 'what' (end-to-end MLOps pipeline orchestration) but has no 'when' clause at all. There is no explicit guidance on when Claude should select this skill, which per the rubric should cap completeness at 3, and the weak 'what' brings it to 2.	2 / 5
Trigger Term Quality	Includes some relevant keywords like 'MLOps', 'pipeline', 'data preparation', 'model deployment', but misses many natural user terms such as 'machine learning', 'ML pipeline', 'model training', 'model serving', 'feature engineering', 'CI/CD', 'experiment tracking', etc.	3 / 5
Distinctiveness Conflict Risk	The term 'MLOps' provides some domain specificity, but 'data preparation' and 'model deployment' are broad enough to overlap with data engineering skills, deployment skills, or general ML skills. The lack of concrete boundaries increases conflict risk.	3 / 5
	Total	10 / 20 Passed

Validation

90%

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 — 10 / 11 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	10 / 11 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills/skills/ml-pipeline-workflow/SKILL.md
Commit: 57c135b

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.