OpenTrustBench

What is OpenTrustBench?

OpenTrustBench is a trust and security verification platform for autonomous AI agents and MCP (Model Context Protocol) servers. It scans your codebase for vulnerabilities, maps findings to OWASP Agentic AI Top 10 and OWASP LLM Top 10 2025, generates a verifiable Trust Card with a letter grade, and provides a shareable badge for your README.

Local analysis, no OpenTrustBench telemetry. Dependency auditing uses the npm registry. No account required.

Quick Start

# Scan a local folder, MCP server, or Agent Skill
npx @opentrustbench/cli scan ./my-mcp-server

# Scan a GitHub repository
npx @opentrustbench/cli scan https://github.com/owner/repo

# Scan an npm package
npx @opentrustbench/cli scan some-npm-package --npm

# Fail CI when findings meet a severity threshold
npx @opentrustbench/cli scan ./my-mcp-server --fail-on high

# Run the OWASP Agentic Top 10 Adversarial Attack Suite
npx @opentrustbench/cli attack ./my-mcp-server

# Evaluate workflow reliability
npx @opentrustbench/cli eval ./tests/workflow.yaml

Install

What You Get

Grade Badge

Embed your trust score in your README:

[![OpenTrustBench](https://www.opentrustbench.com/r/self-packages-cli.svg)](https://www.opentrustbench.com/r/self-packages-cli.html)

This tells buyers, auditors, and AI hosts that your agent has been verified.

How It Works

1. Scan — Point opentrustbench at a local directory, GitHub repo, or npm package
2. Grade — Receive a Trust Card with a letter grade (A–F) and security score (0–100)
3. Share — Embed your badge and link to the detailed report

Features

• 8 detection rules across shell injection, secret leaks, prompt injection, tool exploitation, and more
• OWASP mapping — Every finding tagged to OWASP Agentic Top 10 (ASI01–ASI10) and LLM Top 10 2025 (LLM01–LLM10)
• Permission manifest — Automatic extraction and scoping of agent permissions
• Dependency audit — npm audit integration for lockfile-based vulnerability detection
• SARIF output — Native GitHub Code Scanning integration
• CI integration — --fail-on severity gate for pipeline enforcement
• GitHub Action — Drop-in composite action for workflows
• GitHub URL detection — Scan any public repo by URL
• Compliance helpers — Permission manifests and audit trails to attach to an EU AI Act review (evidence input, not a certification)

What the scanner is (and isn't)

• 8-rule static suite, OWASP-mapped (Agentic ASI01–ASI10, LLM LLM01–LLM10). Regex-based today — no AST yet.
• attack is static-heuristic: it re-analyzes scan findings + permissions. No payloads execute.
• eval is simulation mode: suites are parsed and validated; nothing runs, costs/durations stay 0.
• Not a certification. The Trust Card is a CI-grade credential, not a pentest or legal verdict.

Expert reviews

Need a human pass over your results? Request a review — manual result triage plus a remediation plan, scoped per target. The CLI stays the product and is free forever.

Monorepo Structure

packages/
  core/     — Detection engines, OWASP rules, trust scoring, attack analysis
  cli/      — Terminal CLI with SARIF export and badge generation
  action/   — GitHub Action for CI/CD pipelines
web/        — Landing page and interactive demo
badges/     — Shareable grade badge SVGs
examples/   — Vulnerable and secure test fixtures
scripts/    — Verification and demo scripts
docs/       — Research, registry report, and versioned roadmap (ROADMAP.md)

Build & Test

npm install          # installs + builds via prepare
npm test             # build + 28 unit tests
node scripts/verify-demos.mjs  # end-to-end smoke checks

Standards

• OWASP Agentic AI Top 10 (ASI01–ASI10) — Published December 2025
• OWASP LLM Top 10 2025 (LLM01–LLM10)
• EU AI Act Article 50 transparency obligations (enforceable August 2, 2026)
• SARIF 2.1.0 for GitHub Code Scanning integration

License

Apache 2.0 © 2026 Eulogik