OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Security & Safety

Veritensor

Veritensor is a Python static-analysis security scanner for AI models, notebooks, datasets, dependencies and RAG documents. Its documented checks include potential remote-code-execution risks, data poisoning, personally identifiable information, prompt injections, leaked notebook secrets and malicious notebook content.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Veritensor — OpenViST project profile in AI Security & Safety.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Veritensor when reviewing AI artifacts or documents before they enter an execution environment or RAG pipeline, or when adding documented scanning checks to a development workflow.

Who it is for

  • Machine-learning and AI teams reviewing models, datasets or notebooks before use.
  • RAG developers inspecting documents before ingestion into a retrieval pipeline.
  • Security and DevSecOps teams considering artifact checks in development or release workflows.

Documented uses

  • Scan model and notebook files for potential execution risks and other threats.
  • Check datasets and RAG documents for poisoning patterns, prompt injections and PII.
  • Inspect project dependencies for typosquatting and known vulnerabilities.
  • Create a JSON data manifest for an artifact folder.
Before you choose

Requirements & limits

The project is open source under the Apache-2.0 license. The repository links PyPI, Docker Hub and a Hugging Face Space; the README also documents CLI installation through pip.

  • The core pip package is listed for models, notebooks and dependencies; RAG document scanning and ML-based PII detection are listed as optional extras.
  • Dataset scans sample 10,000 rows by default; the README provides a --full-scan option.
  • The README documents Python integrations for LangChain, LlamaIndex and ChromaDB, separate from its CLI and container distribution routes.
  • Docker is recommended in the README for CI/CD, but runtime configuration requirements are not specified there.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Recursively scan a local AI project

Run the documented CLI scan over a project directory for supported threats.

Before you start

  • Install the Veritensor core package with pip install veritensor.
  • Have a local project directory to scan.

Steps

  1. Install the core package with the command below.

    Documented command
    pip install veritensor

  2. Run veritensor scan ./my-rag-project --recursive --jobs 4, substituting the path to your project.

Expected outcome

Veritensor scans the directory recursively using four CPU cores and presents scan results.

Guide limitations

  • The core installation is described as covering models, notebooks and dependencies; document scanning is listed under the RAG extra.
  • The command scans supported threats, but the README does not specify a complete inventory of formats covered by this particular command.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
arsbr/Veritensor · Official homepage title · retained in its source language
How we review sourcesSuggest a correction