🤖 AI Analysis

Final verdict: SAFE

The package shows minimal risk indicators, with no network calls, shell executions, or signs of obfuscation or credential harvesting. The metadata risk slightly increases due to the maintainer's limited presence on PyPI.

No network calls
No shell execution patterns
Maintainer has only one package

Per-check LLM notes

Network: No network calls detected, which is normal if the package does not require internet access.
Shell: No shell execution patterns detected, indicating no direct system command execution from the package.
Obfuscation: No obfuscation patterns detected, indicating low risk of malicious intent.
Credentials: No credential harvesting patterns detected, suggesting no risk of secret theft.
Metadata: The maintainer has only one package and lacks PyPI classifiers, indicating low effort or a new account.

🔬 Heuristic Checks

✓ Outbound Network Calls

No suspicious network call patterns found

✓ Code Obfuscation

No obfuscation patterns detected

✓ Shell / Subprocess Execution

No shell execution patterns detected

✓ Credential Harvesting

No credential harvesting patterns detected

✓ Typosquatting

No typosquatting candidates detected

✓ Registered Email Domain

Email domain looks legitimate: lsi.uned.es

✓ Suspicious Page Links

All external links appear legitimate

✓ Git Repository History

Repository UNEDLENAR/PyEvALL appears legitimate

⚠ Maintainer History score 4.0

2 maintainer concern(s) found

Author "JORGE CARRILLO-DE-ALBORNOZ" appears to have only 1 package on PyPI (new or inactive account)
Package has no PyPI classifiers (low effort / metadata quality)

✓ Known CVE Vulnerabilities

No known vulnerabilities found in OSV database.

💡 AI App Starter Prompt

Use this prompt to build a project with PyEvALL

Create a mini-application named 'EvalMaster' that leverages the PyEvALL Python package to evaluate different types of machine learning models. The application should allow users to upload their datasets and select from a variety of evaluation contexts such as classification, ranking, and LeWiDi. Users should also be able to choose which specific metrics they want to assess their model against within each context. Here’s a step-by-step guide on what the application should do:

1. **Setup and Interface**: Design a user-friendly interface where users can upload their dataset and specify the type of evaluation they wish to perform (classification, ranking, or LeWiDi).
2. **Model Input**: Allow users to input the trained model they want to evaluate. Ensure that the application supports common model formats.
3. **Evaluation Context Selection**: Provide options for users to select the evaluation context (classification, ranking, LeWiDi) and then choose specific metrics relevant to that context from a dropdown menu.
4. **Execution and Results**: Once the user has selected their preferences, the application should use PyEvALL to execute the chosen evaluations and display the results in an easily understandable format.
5. **Visualization**: Implement basic visualization tools to help users interpret the evaluation results more effectively.
6. **Documentation and Help**: Include comprehensive documentation and a help section explaining how to use the application, what each metric means, and how PyEvALL works under the hood.

**Features to Consider**:
- Support for multiple file formats for dataset uploads.
- Pre-built examples for quick testing without needing to upload data.
- Integration with popular machine learning libraries like scikit-learn, TensorFlow, and PyTorch.
- Option to save and share evaluation results.

Utilize PyEvALL's core functionalities by invoking its methods for evaluating classification accuracy, ranking precision, and handling disagreements in LeWiDi scenarios. Ensure that your application demonstrates the versatility of PyEvALL across these different evaluation contexts.