model-export-helper

Model Export Helper - Auto-activating skill for ML Deployment. Triggers on: model export helper, model export helper Part of the ML Deployment skill category.

Install with Tessl CLI

npx tessl i github:jeremylongshore/claude-code-plugins-plus-skills --skill model-export-helper

What are skills?

1.00x

Quality

Does it follow best practices?

Impact

83%

1.00x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./planned-skills/generated/08-ml-deployment/model-export-helper/SKILL.md

SKILL.md

Review

Evals

Evaluation results

83%

Packaging a Trained Classifier for Handoff to the Deployment Team

Step-by-step export guidance with validation

Criteria

Without context

With context

Sequential steps in guide

100%

Error handling in export script

20%

Logging in export script

100%

Configurable output path

Validation step present

100%

Validation uses test input

100%

Exported format documented

75%

100%

Dependency specification

100%

Separation of concerns

62%

100%

No hardcoded credentials

100%

Guide covers all key stages

100%

Without context: $0.6559 · 3m 12s · 28 turns · 31 in / 10,482 out tokens

With context: $0.7192 · 3m 9s · 32 turns · 32 in / 11,279 out tokens

86%

-6%

Deploying a Fraud Detection Model as a Production REST API

Production-ready model serving configuration

Criteria

Without context

With context

No hardcoded config

100%

Health check endpoint

100%

Model versioning metadata

11%

Request/response logging

100%

Input validation

100%

Error handling in predict path

100%

44%

Dependency file present

100%

Containerization config

100%

Model loaded once at startup

100%

Serving guide structured

100%

No credentials in code

100%

Without context: $0.6488 · 2m 38s · 34 turns · 35 in / 9,753 out tokens

With context: $0.6008 · 2m 36s · 30 turns · 28 in / 8,818 out tokens

82%

Setting Up Ongoing Health Monitoring for a Deployed Credit Scoring Model

MLOps monitoring with validation against standards

Criteria

Without context

With context

Baseline reference stored

100%

Drift detection logic present

100%

Configurable thresholds

70%

100%

Sequential setup guide

100%

Alert or report output

100%

Error handling in monitor

44%

55%

Logging present

62%

37%

Performance metric tracked

100%

Validation against schema

22%

No hardcoded paths or secrets

87%

75%

Reproducible environment

100%

Without context: $0.6666 · 3m 16s · 25 turns · 25 in / 12,738 out tokens

With context: $0.8966 · 4m · 38 turns · 37 in / 13,651 out tokens

Evaluated: 6 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Packaging a Trained Classifier for Handoff to the Deployment Team Deploying a Fraud Detection Model as a Production REST API Setting Up Ongoing Health Monitoring for a Deployed Credit Scoring 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.