OWASP-mapped · Agentic + LLM Top 10

A trust score for every AI agent.

AgentTrust is a free, local-first security scanner for AI agents and MCP servers. Eight static rules, a permission manifest, and a Trust Card graded A–F — in seconds, with zero data leaving your machine.

npx @eulogik/agenttrust scan . View on GitHub

Free forever · No account · Open source (Apache-2.0) · Install in 60 seconds

50
public MCP servers scanned — browse the registry
70.2
average score (C) across the ecosystem
36%
grade D or F — findings need triage, not panic
0
bytes of your code ever transmitted
terminal — sample output, secure fixture

How it works

Three steps to a verifiable grade.

01 — SCAN

Point at anything

A local directory, a GitHub URL, or an npm package. Recursive detection, eight rules, permission extraction, dependency audit. Seconds, locally.

02 — GRADE

Get a Trust Card

Letter grade A–F, 0–100 score, OWASP-mapped findings with file:line, plus SARIF for your Security tab and a --fail-on CI gate.

03 — PROVE

Share the evidence

Embed a bound badge that links to a public report page — grade image and evidence travel together. See the live registry.

The registry · 50 servers, re-scanned weekly

Who's passing, who's not.

Independently scanned snapshots — not self-reports. Full evidence behind every grade.

THE PATTERN

Scope predicts the grade.

Every A-grade server pairs zero findings with minimal scope. All D/F servers are excessive-scope except two. How scoring works · Explore all 50 →

Install

Running in 60 seconds.

npm install -g @eulogik/agenttrust
agenttrust scan ./my-mcp-server --quiet

No install? Run it directly with npx @eulogik/agenttrust scan . From source: git clone https://github.com/eulogik/AgentTrust.git && cd AgentTrust && npm install. Enforce it in CI:

agenttrust scan . --fail-on high --quiet --output-dir ./trust

Or drop in the GitHub Action — scan plus SARIF upload to your Security tab.

Pricing

The CLI is the product. It's free.

CLI

$0, forever

Unlimited local scans, Trust Cards, bound badges, SARIF, CI gate. Open source, zero telemetry.

EXPERT REVIEW

Scoped per target

Manual triage of your results plus a remediation plan, on request. Contact us.

ENTERPRISE

Custom

Custom rules, evidence helpers, priority support. Contact us.

FAQ

Asked, answered.

What is AgentTrust?

AgentTrust is a free, local-first security scanner for AI agents and MCP servers. It runs 8 OWASP-mapped static rules, extracts a permission manifest, and issues a Trust Card graded A–F — with SARIF output for CI. It answers: can I trust this agent, what may it do, and can I prove it?

How do I scan an MCP server?

Point the CLI at a local directory, a GitHub URL, or an npm package. You get a Trust Card, SARIF and Markdown reports in seconds. Nothing leaves your machine — GitHub targets are cloned to a temp directory locally.

What is a Trust Card?

A portable, machine-readable credential (agenttrust/trust-card/v1): letter grade, 0–100 score, findings with file:line evidence, permission scope, provenance signals, and inferred host compatibility. Bound badges link each grade to its public evidence page.

Is AgentTrust a certification?

No. AgentTrust is a static scanner, not a certification body. Attack analysis is heuristic (no payloads execute), workflow eval is simulated (nothing runs), and Trust Cards are evidence input for your own review — never a compliance verdict.

Does my code leave my machine?

No. Zero telemetry, zero retention. The scanner reads files, computes locally, and writes reports to a directory you choose.

What do the grades mean?

A (90+) through F (below 40), weighted across security findings, permission scope, provenance, reliability and stability. Across 50 public MCP servers the average is 70.2 (C); 36% grade D or F. Full scoring methodology.

Why does the AgentTrust repo itself grade F?

Because it contains its own test ammunition — and a scanner that didn't flag it would be the real scandal. The F comes from three intentional sources: (1) examples/vulnerable-mcp-server, a deliberately vulnerable fixture; (2) our unit tests, which must contain eval(, exec( and transferFunds( strings to prove the detector catches them; (3) our automation scripts, which legitimately shell out to git and npm with hardcoded local arguments. The shipped product (packages/cli) self-scans at A (90/100) — that is what the README badge grades. Rule of thumb: grade the artifact you ship, not the monorepo that tests it.

Ship agents you can defend.

One command. Thirty seconds. A grade you can show your auditor.

npx @eulogik/agenttrust scan . Browse the registry