Skip to main content
Glama

Live Web App CI Status License Python Version FastMCP

Launch Free Web Scanner Python SDK Claude MCP Server Privacy Architecture Validation Report Reporting Guide


Architecture & Threat Vectors


Related MCP server: relayshield-mcp

30-Second Quickstart

Option A: Use Instantly Online (Zero Installation)

Visit the public web scanner directly from any device:
https://sasindudilshanranwadana.github.io/aussie-shield/

Option B: Deep Multilingual Analysis (Opt-in)

The public web app defaults to Private Check: message text stays in the browser and only local indicators run. Users can explicitly enable Deep Analysis to send up to 2,000 characters over HTTPS to the project Cloudflare Worker, where Workers AI returns a structured multilingual triage response. It is not a verification or safety guarantee.

See worker/README.md for the Worker deployment, consent, rate-limit, and data-disclosure contract.

Option C: Terminal CLI (Python / uv)

# Run instantly with uv
uvx aussie-shield check-message "Your ATO tax refund of $1,200 is pending: bit.ly/myato-secure"

# Or install via pip
pip install aussie-shield

Core Capabilities


CLI Showcase

1. Evaluate Suspicious URLs

aussie-shield check-url "https://myg0v-verify.com/login"
[HIGH RISK] HIGH RISK — Domain resembles myGov (possible phishing); Domain is 3 day(s) old

  URL:         https://myg0v-verify.com/login
  Domain:      myg0v-verify.com
  Domain age:  3 day(s)
  Flagged by:  Impersonation Rule Engine, RDAP Bootstrap

  Action: Do NOT visit this URL or enter credentials.
  Report: https://portal.scamwatch.gov.au/report-a-scam/

2. Check SMS / Email Texts

aussie-shield check-message "Final notice: Unpaid tax debt of $3,400 with ATO. Purchase $500 iTunes gift cards immediately or warrant will be issued."
[CRITICAL RISK] Payment demanded via gift cards (iTunes, Google Play, Steam) detected (confidence: 95%)

  Matched indicators:
    - gift card
    - ato
    - debt
    - warrant

  What to do:
    1. No government agency, bank, or legitimate business will EVER ask for payment via gift cards
    2. This is ALWAYS a scam
    3. Report to Scamwatch immediately (https://portal.scamwatch.gov.au/report-a-scam/)

3. Verify Business ABN & Invoice Alignment

aussie-shield check-abn 51824753556 --name "Amazon Australia"
[ACTIVE] ABN 51 824 753 556 — [Registered Entity] (Australian Private Company) | Name: MISMATCH

  ABN:              51 824 753 556
  Registered Name:  [Entity Name on Register]
  Status:           Active
  GST Registered:   Yes
  Name Match:       MISMATCH

  Risk Flags:
    - Name mismatch: Claimed 'Amazon Australia' but registered to '[Entity Name]'. Possible invoice redirection fraud.

Claude Desktop / MCP Server

Integrate aussie-shield directly into Claude Desktop or any MCP client for autonomous scam detection during conversation:

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "aussie-shield": {
      "command": "aussie-shield",
      "args": ["serve"],
      "env": {
        "AUSSIE_SHIELD_GOOGLE_KEY": "your-google-safe-browsing-key",
        "AUSSIE_SHIELD_ABR_GUID": "your-abr-guid",
        "AUSSIE_SHIELD_ABUSEIPDB_KEY": "your-abuseipdb-key"
      }
    }
  }
}

Exposed MCP Tools

  • check_url(url) — Multi-feed threat lookup + domain age

  • check_message(text) — 20+ AU scam patterns (100% local)

  • check_abn(abn, claimed_name) — ABR register lookup & name reconciliation

  • check_phone(number) — Australian E.164 and 190x risk analysis

  • generate_report(...) — Formatted Scamwatch/ReportCyber submission draft

  • get_scam_briefing(scam_type) — Tactical response briefings


Python SDK

from aussie_shield.checkers.message_checker import check_message
from aussie_shield.checkers.url_checker import check_url
from aussie_shield.checkers.abn_checker import check_abn
from aussie_shield.checkers.phone_checker import check_phone

# Message scan (Zero Network)
msg_res = check_message("AusPost: Parcel held at depot. Pay $3.50 fee at bit.ly/auspost-fee")
print(msg_res.risk_level)       # 'HIGH'
print(msg_res.scam_type)        # 'AusPost / Delivery Fee Scam'
print(msg_res.what_to_do)       # List of remediation steps

# Domain scan
url_res = check_url("https://myg0v-auth.com/login")
print(url_res.risk_level)       # 'HIGH'
print(url_res.impersonation_flags) # ["Domain resembles myGov (possible phishing)"]

# ABN verification
abn_res = check_abn("51824753556", claimed_name="Amazon Australia")
print(abn_res.valid_format)     # True
print(abn_res.claimed_name_match) # False (triggers invoice alert)

Privacy Architecture

Engine Tool

Data Sent Externally

Destination

Protocol / Privacy Safeguard

check_message Private Check

None

Local Machine Only

Default browser-only local indicators

check_message Deep Analysis

Message text after explicit consent

Cloudflare Worker / Workers AI

Optional multilingual triage; message text is not persisted by the Worker

check_url local allowlist / impersonation

None

Local Machine Only

Known-safe domains and brand indicators are evaluated locally

check_url Google Safe Browsing v4 / PhishTank

Full URL when the lookup runs

Third-party threat feeds

Use only where the user accepts disclosure to the selected feed

check_url RDAP / URLScan.io

Domain only

Registry / URLScan.io

Domain string only; paths and tokens are stripped

check_url AbuseIPDB

IP address only, when an IP URL is supplied

AbuseIPDB

Requires a user-provided API key

check_abn

ABN Number

ABR Public Register

Public business identifier

check_phone

None

Local Machine Only

100% Offline Regex & E.164 Engine

generate_report

None

Local Machine Only

Markdown generated locally


Incident Response: If You Have Been Scammed

  1. Immediate Financial Freeze — Call your financial institution's fraud hotline immediately within 24 hours to attempt transaction recall.

  2. Report to Scamwatch (ACCC / NASC)portal.scamwatch.gov.au/report-a-scam

  3. Report to ACSC ReportCybercyber.gov.au/report-and-recover/report if passwords, credentials, or identity documents were compromised.

  4. National Identity Support (IDCARE) — Call 1800 595 160 (idcare.org) for identity recovery planning.

See the full Incident Reporting Guide for direct numbers to major Australian banks (CommBank, ANZ, Westpac, NAB, ING).


aussie-shield provides indicative risk scoring based on algorithmic pattern matching and threat intelligence feeds. It does not constitute legal, financial, or law enforcement advice.

This project is independently developed and maintained under the MIT License. It is not affiliated with, operated by, or endorsed by the Australian Government, ACCC, ACSC, or the Australian Taxation Office. See DISCLAIMER.md for full terms.


Engineered with precision for Australia by @sasindudilshanranwadana
Master of Information Technology (Cybersecurity) · Charles Darwin University

GitHub Stars GitHub Forks


Support & Community

If you find this project useful, you can support continuous development:

Buy Me A Coffee

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

No tool schema history has been recorded yet.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

  • A
    license
    A
    quality
    F
    maintenance
    AI-powered threat verification across 6 channels: phone numbers, URLs, text messages, emails, documents, and QR codes. Analyzes threats against 10M+ intelligence records from FTC, FCC, URLhaus, ThreatFox, and community reports, returning risk scores, verdicts, and detailed signals.
    10
    1
    MIT
  • A
    license
    A
    quality
    A
    maintenance
    Security intelligence for AI agents — breach detection, SIM swap, domain lookalikes, OAuth watchlist, and malware scanning. Subscription or x402 PAYG.
    11
    MIT
  • A
    license
    A
    quality
    C
    maintenance
    Scans suspicious messages, URLs, and text for scams inside any MCP-compatible AI assistant. No signup or API key needed for anonymous use.
    1
    57
    MIT
  • F
    license
    Not graded
    quality
    D
    maintenance
    Provides real-time threat intelligence for AI agents, enabling checks on IPs, domains, URLs, hashes, CVEs, prompt-injection payloads, and malicious AI-skill/MCP-tool definitions against a free database of 890K+ IOCs.
    -

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/sasindudilshanranwadana/aussie-shield'

If you have feedback or need assistance with the MCP directory API, please join our Discord server