What transport does Apviso MCP use?
The server uses the `stdio` transport.
@apviso
基本情報
MCP server for interacting with the APVISO AI-powered penetration testing platform from Claude Code, Cursor, Windsurf, Codex, and other MCP-compatible tools.
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"apviso": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@apviso/mcp"
],
"env": {
"APVISO_API_KEY": "apvk_your_key_here"
}
}
}
}Get your current quota usage including subscription tier, credits remaining, and billing period dates. Use this to check available scan credits before starting a new scan.
List all your registered targets (domains). Returns paginated results with verification status.
Get details for a specific target including domain, verification status, and whether authentication is configured.
Register a new target domain for scanning. The target must be verified before it can be scanned. After creating, use get_verification_instructions to see how to verify ownership.
Verify ownership of a target using one of three methods: dns_txt (add a DNS TXT record), file (upload a verification file to /.well-known/penterep-verify.txt), or meta_tag (add a meta tag to your homepage). Use get_verification_instructions first to see the required values.
Get the verification token and step-by-step instructions for all three verification methods (DNS TXT, file upload, meta tag) for a target.
Delete a target. This will fail if the target has any associated scans.
List your scans with optional filtering by status. Returns paginated results including scan status, target info, and timestamps.
Get details for a specific scan including status, target, model preset, timestamps, and whether it's a retest. Use this to check scan progress.
Start a new penetration test scan. IMPORTANT: This creates a billable scan that costs credits. The target must be verified first. Check quota with get_quota before starting. modelPreset controls depth: 'free' uses free credits, 'low' is fastest, 'ultra' is most thorough and expensive.
List findings (vulnerabilities) for a specific scan. Results are paginated and descriptions are truncated — use get_report for full finding details. Filter by severity (critical/high/medium/low/info) or status (open/in_progress/fixed/accepted_risk/false_positive).
Update the status of a finding. Use this to track remediation progress: open → in_progress → fixed, or mark as accepted_risk or false_positive.
Get the full penetration test report for a completed scan. Returns the report as markdown content with complete finding details, evidence, and remediation guidance. Also includes a PDF download URL if available. This is the best tool for getting comprehensive scan results.
List all your scan schedules. Schedules automatically run scans on a recurring basis (daily, weekly, biweekly, or monthly). Requires Business or Enterprise tier.
Get details for a specific scan schedule including frequency, next run time, and configuration.
Create a recurring scan schedule for a verified target. Requires Business or Enterprise tier. Scans will run automatically at the specified time and frequency.
Update an existing scan schedule. You can change the frequency, timing, model preset, or enable/disable it.
Delete a scan schedule. This stops all future scheduled scans for this target.
Apviso MCP is a Model Context Protocol server that connects AI coding assistants (Claude Code, Cursor, Windsurf, Codex) to the APVISO AI-powered penetration testing platform. It enables developers to register targets, launch scans, view findings, and manage security testing directly from their IDE.
Obtain an API key from your APVISO dashboard, then configure the server using Claude Code (claude mcp add) or by editing .mcp.json / ~/.claude.json with the required APVISO_API_KEY environment variable. The server runs via npx -y @apviso/mcp as a stdio transport.
The server uses the `stdio` transport.
`APVISO_API_KEY` is required; it must be an API key starting with `apvk_`.
Inside Claude Code, run `/mcp` to check the server status.
Yes, set the optional `APVISO_API_URL` environment variable; it defaults to `https://apviso.com/api`.
Yes, you can configure it in any MCP-compatible client by setting the command and environment variables as shown in the manual config section.

Verified cyber-physical AI tools. 25 MCP tools, x402-payable on Base.

Leadpages MCP is a remote MCP server available at https://mcp.leadpages.com.
Gives AI agents public URLs (tunnels) for localhost, live HTTP traffic inspection, snapshot publishing, and access control. Part of LocalCan, the ngrok alternative for Mac, Windows and Linux. Free plan.
Modelglass turns "which model should I use?" into a lookup instead of a guess. Seven tools give your agent live pricing across image, language, video, and audio models. Sourced and versioned, not scraped once and forgott

Nebelus is an enterprise agentic-AI platform for regulated industries. This MCP server exposes the Nebelus Construction API: ~48 tools that give an MCP client full agent build parity — create/update/probe agents, edit ag
コメント