flow-nexus-neural
Train and deploy neural networks in distributed E2B sandboxes with Flow Nexus
Install with Tessl CLI
npx tessl i github:majiayu000/claude-skill-registry-data --skill flow-nexus-neural60
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Discovery
32%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 a specific technical domain (neural network training/deployment) and mentions proprietary tools (E2B, Flow Nexus), but suffers from missing trigger guidance and incomplete keyword coverage. The lack of a 'Use when...' clause significantly limits Claude's ability to know when to select this skill over other ML-related options.
Suggestions
Add a 'Use when...' clause with explicit triggers like 'Use when the user wants to train ML models, deploy neural networks, or mentions E2B, Flow Nexus, or distributed model training'
Include common user-facing terms like 'machine learning', 'ML', 'deep learning', 'model training', 'GPU training' alongside the technical product names
Expand specific capabilities beyond just 'train and deploy' - mention supported frameworks, scaling options, or monitoring features if applicable
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (neural networks, distributed sandboxes) and two actions (train, deploy), but lacks comprehensive detail about specific capabilities like model types, training configurations, or deployment options. | 2 / 3 |
Completeness | Describes what it does (train/deploy neural networks) but completely lacks a 'Use when...' clause or any explicit trigger guidance for when Claude should select this skill. | 1 / 3 |
Trigger Term Quality | Includes some relevant technical terms ('neural networks', 'distributed', 'sandboxes', 'train', 'deploy') but 'E2B sandboxes' and 'Flow Nexus' are product-specific jargon users may not naturally use. Missing common variations like 'ML', 'machine learning', 'deep learning', 'model training'. | 2 / 3 |
Distinctiveness Conflict Risk | The combination of 'E2B sandboxes' and 'Flow Nexus' provides some distinctiveness, but 'neural networks' and 'train/deploy' could overlap with other ML-related skills. The product names help but aren't sufficient without clearer scope boundaries. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides excellent actionability with comprehensive, executable code examples covering neural network training, distributed clusters, and model management. However, it suffers from verbosity with repetitive examples and patterns that could be consolidated, and lacks explicit validation checkpoints in multi-step distributed training workflows. The monolithic structure would benefit from splitting into focused reference files.
Suggestions
Add explicit validation steps in distributed training workflows (e.g., 'Verify cluster status shows initializing before deploying nodes', 'Check all nodes are active before starting training')
Consolidate architecture patterns into a separate ARCHITECTURES.md file and reference it, keeping only one representative example in the main skill
Remove redundant 'Best for:' descriptions in architecture patterns - Claude can infer appropriate use cases from the architecture type
Add error handling guidance inline with workflows (e.g., what to do if cluster_init returns an error status)
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is comprehensive but overly verbose with many similar examples that could be consolidated. Architecture patterns section repeats information already shown in examples, and some explanations (like 'Best for:' descriptions) add minimal value for Claude. | 2 / 3 |
Actionability | Excellent actionability with fully executable JavaScript code examples, complete with realistic parameters, expected response formats, and copy-paste ready MCP tool calls. Every capability is demonstrated with concrete, runnable code. | 3 / 3 |
Workflow Clarity | Multi-step workflows like distributed training show the sequence (init cluster → deploy nodes → connect → train → monitor) but lack explicit validation checkpoints. No guidance on verifying cluster initialization succeeded before deploying nodes, or checking node health before starting training. | 2 / 3 |
Progressive Disclosure | References external docs and related skills at the end, but the main content is a monolithic 500+ line document. Architecture patterns, common use cases, and troubleshooting could be split into separate files with clear navigation from a concise overview. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (739 lines); consider splitting into references/ and linking | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- 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.