// MCP TOOL REFERENCE

ARC SECURITY TOOLS

14 tools from the Arc Security MCP Server, categorised by risk level.

// ALL 14 ARC SECURITY TOOLS

READ 13 tools

Read analyze_skill_code Analyze skill source code for dangerous patterns. Scans code against Arc Security's pattern database b... Read analyze_skill_intent Analyze a SKILL.md file for intent mismatches and semantic threats. Uses AI to read the skill's instru... Read check_skill_safety Check if a ClawHub skill or MCP server is safe to install. First checks Arc Security's curated databas... Read get_attack_class_info Get detailed information about a known agent attack class. Arc Security has documented 22 distinct att... Read get_owasp_mapping Get the mapping between Arc Security's 22 attack classes and OWASP Agentic AI Top 10. Shows how each o... Read get_session_alerts Get all alerts for a running monitoring session. Args: session_token: Token from monitor_start() Read get_threat_landscape Get the current AI agent security threat landscape summary. Returns key statistics, active threats, an... Read list_dangerous_patterns List all known dangerous code patterns that Arc Security checks for. Returns the full pattern database... Read monitor_end End monitoring and get a session report. Args: session_token: Token from monitor_start() Read monitor_event Report a tool call, file access, or network request for monitoring. Args: session_token: Token... Read monitor_start Start monitoring an AI agent session. Returns a session token for tracking. Args: session_id: ... Read scan_skill_realtime Fetch a skill from ClawHub and perform a full security scan in real-time. Downloads the skill's SKILL.... Read security_checklist Get a security checklist for a specific type of skill. Returns a tailored checklist of security requir...

WRITE 1 tools

Write set_monitor_policy Set behavioral policy for a skill (allow/deny rules for runtime monitoring). Args: skill_name:...

// RUNNING THIS SERVER

The managed route: connect Arc Security through the PolicyLayer gateway — every tool call above is checked against your policy before it runs, with a full audit log.

DIRECT INSTALL (UNMANAGED) npx -y arc-security-mcp

// FAQ

How many tools does the Arc Security MCP server have? +

The Arc Security MCP server exposes 14 tools across 2 categories: Read, Write.

How do I enforce policies on Arc Security tools? +

Route the Arc Security server through the PolicyLayer gateway. Define allow, deny, or approval rules per tool in the dashboard — they are enforced on every call before it reaches the server.

What risk categories do Arc Security tools fall into? +

Arc Security tools are categorised as Read (13), Write (1). Each category has a recommended default policy.

Let agents act without letting them run wild.

Route your MCP servers through PolicyLayer and every tool call is checked against your policy before it runs — allow, deny, or require approval. Per-identity grants. Full audit log. Live in minutes.

SECURE YOUR MCP →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.