flow-nexus-neural

Train and deploy neural networks in distributed E2B sandboxes with Flow Nexus

2.07x

Quality

37%

Does it follow best practices?

Impact

79%

2.07x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./.agents/skills/flow-nexus-neural/SKILL.md

Evaluation results

41%

24%

Credit Card Fraud Detection Prototype

Single-node neural training configuration

Criteria

Without context

With context

Correct train function

Valid architecture type

100%

Valid tier for prototyping

100%

Training optimizer

100%

Training epochs

100%

Training batch_size

40%

100%

Training learning_rate

40%

100%

Divergent enabled

Valid divergent pattern

Divergent factor

Training status check

Correct predict function

96%

Large-Scale Image Recognition Cluster Setup

Distributed cluster setup and orchestration

Criteria

Without context

With context

Correct cluster init function

50%

75%

Valid cluster topology

100%

Valid consensus mechanism

100%

daaEnabled set

100%

wasmOptimization set

100%

Parameter server node

100%

Worker node

100%

Aggregator node

100%

Cluster connect call

100%

Distributed train function

55%

77%

Valid optimizer in distributed train

100%

Cluster status monitoring

100%

Cluster terminate call

100%

93%

Privacy-Preserving Medical Diagnosis Model

Federated learning and model lifecycle management

Criteria

Without context

With context

Federated flag set

100%

aggregation_rounds configured

100%

min_nodes_per_round configured

100%

proof-of-learning consensus

100%

Worker autonomy set

100%

daaEnabled for federated cluster

100%

Status monitoring call

40%

100%

Performance benchmark

100%

Validation workflow

100%

Publish template function

100%

Publish category set

83%

100%

Repository: ruvnet/ruflo
Commit: 398f7c2

Evaluated: about 1 month ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Credit Card Fraud Detection Prototype Large-Scale Image Recognition Cluster Setup Privacy-Preserving Medical Diagnosis Model

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.