// SCAN REPORT

Pentest Ai

52 tools. 32 can modify or destroy data without limits.

32 write tools that can modify data. Rate limits recommended.

Last updated: 7 Aug 2026

32 can modify or destroy data

20 read-only

52 tools total

Community server · catalogue entry checked 07/08/2026

How to control Pentest Ai ↓

TOOL MAP

What Pentest Ai exposes to your agents

Read (20) Write / Execute (32) Destructive / Financial (0)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

High Risk

The most dangerous Pentest Ai tools

Execute · High run_probe The tool appears designed to execute security testing probes/scans against systems. Execute · High run_recon Although the description is empty, the tool name 'run_recon' combined with the server's explicit purpose (autonomous pentests using real security tools) indicates this executes reconnaissance operations. Execute · Critical run_tool This is an Execute category tool because it runs code/tools whose side effects (network scanning, exploitation attempts, credential testing, vulnerability verification) depend entirely on the arguments provided. Execute · High start_campaign This tool initiates active pentesting engagements—it executes security scanning and attack operations whose effects (network traffic, system probing, exploit attempts) depend on the targets specified and campaign configu

32 of Pentest Ai's 52 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Pentest Ai

PolicyLayer is an MCP gateway — it sits between your AI agents and Pentest Ai, and nothing reaches the server without passing your rules. These are the rules we recommend:

Rate limit write operations

{
  "close_engagement": {
    "limits": [
      {
        "counter": "close_engagement_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "discover_attack_chains": {
    "limits": [
      {
        "counter": "discover_attack_chains_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Pentest Ai — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON PENTEST AI →

Instant setup, no code required.

FULL CATALOGUE

All 52 Pentest Ai tools

EXECUTE 29 tools

Execute authenticated_scan authenticated_scan Execute browser_inspect Inspect a URL with the headless browser. Execute builtin_scan builtin_scan Execute ensure_tools_installed ensure_tools_installed Execute http_request http_request Execute kill_process Terminate a running tool subprocess by PID. Execute plan_tools plan_tools Execute poll_oob poll_oob Execute prove_attack_chain prove_attack_chain Execute resume_engagement resume_engagement Execute run_probe run_probe Execute run_recon run_recon Execute run_tool Run a specific security tool against a target. Execute scan_dns_builtin Perform DNS enumeration (built-in). Execute scan_ports_builtin Scan common ports on a target (built-in, no nmap required). Execute select_agent select_agent Execute set_intensity Change scan intensity (stealth, normal, aggressive) mid-engagement. Execute start_campaign Start a multi-target campaign. Creates one engagement per target. Execute start_engagement start_engagement Execute test_active_directory test_active_directory Execute test_api_security Run API security testing (REST + GraphQL) following OWASP API Top 10. Execute test_cloud test_cloud Execute test_credentials Run authentication testing (default creds, password spray, MFA bypass). Execute test_mobile Run mobile app security testing (Android or iOS). Execute test_privesc Run privilege escalation enumeration on a compromised host. Execute test_social_engineering Run a social engineering assessment (phishing simulation, OSINT, DMARC audit). Execute test_vulnerabilities Run vulnerability scanning (Nuclei + nikto; nuclei CVE templates are the primary known-CVE path). Execute test_web_app test_web_app Execute test_wireless Run wireless security assessment (WiFi + Bluetooth). Returns

WRITE 3 tools

Write close_engagement Close an engagement and mark it as completed. Write generate_detection_rules generate_detection_rules Write generate_report generate_report

READ 20 tools

Read discover_attack_chains Discover attack chains from existing findings. Read get_attack_chains Get discovered attack chains for an engagement. Read get_campaign_summary Get aggregated summary across all engagements in a campaign. Read get_config Get current pentest-ai configuration (secrets masked). Read get_engagement_status Get the current status of a pentest engagement. Read get_engagement_summary Get a summary of an engagement including finding counts, chains, and rules. Read get_evidence get_evidence Read get_findings get_findings Read health health Read list_engagements List all pentest engagements, optionally filtered by status. Read list_plugins List installed YAML plugins from ~/.pentest-ai/plugins/. Read list_probes list_probes Read list_processes List running tool subprocesses tracked by the engine. Read list_tools List all available security tools, optionally filtered by category. Read query_compliance Query compliance mapping for an engagement's findings. Read scan_headers_builtin Analyze HTTP security headers (built-in). Read scan_paths_builtin Scan for common sensitive paths (built-in). Read scan_secrets_builtin Scan HTTP responses for leaked secrets and credentials (built-in). Read scan_ssl_builtin Check SSL/TLS configuration (built-in). Read validate_finding validate_finding

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Pentest Ai

How do I prevent bulk modifications through Pentest Ai? +

The Pentest Ai server has 3 write tools including close_engagement, generate_detection_rules, generate_report. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Pentest Ai.

How many tools does the Pentest Ai MCP server expose? +

52 tools across 3 categories: Execute, Read, Write. 20 are read-only. 32 can modify, create, or delete data.

How do I enforce a policy on Pentest Ai? +

Register the Pentest Ai MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Pentest Ai tool call.

Deterministic rules across all 52 Pentest Ai tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON PENTEST AI →

Instant setup, no code required.

52 Pentest Ai tools catalogued and risk-classified — across an index of 46,500+ MCP servers.

// WHERE THIS COMES FROM

These policies come from Pentest Ai's registry record.

The record behind this page: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.

Teams ship this data inside their own products. See what a licence covers →