Master Metasploit engagement orchestrator. Use when the user says "start testing", "run recon", "exploit targets", or gives broad multi-phase instructions. Drives the phase/gate/subgate workflow and coordinates specialized subagents.
Metasploit exploitation specialist. Use for vulnerability checks, exploit execution, handler setup, and session verification. Only operates against targets approved by the orchestrator and within ROE.
Metasploit post-exploitation specialist. Use for session interaction, running post modules, credential harvesting, loot collection, and session cleanup.
Reconnaissance and scanning specialist. Runs nmap, enumerates services, queries the MSF database, searches modules, and profiles targets. NEVER exploits -- only observes and reports.
Read-only security reviewer for Metasploit engagement plans. Use before any destructive operation to validate targets, modules, and ROE compliance. Enforces G4 hard gate (SG4.1-reviewer-gate). Does not execute any actions.
Guides AI and LLM application security testing including prompt injection, system prompt extraction, tool/function abuse, MCP server testing, and data exfiltration via LLM features. Use for chatbots, agents, and AI-integrated apps.
Guides binary exploitation including stack overflow, ROP, format strings, heap basics, pwntools, and GDB analysis. Use when engagement ROE explicitly authorizes custom exploit development beyond Metasploit modules.
Guides blockchain and smart contract security testing with reentrancy, flash loan, and access control analysis using Slither and EVM tooling. Use when auditing Solidity contracts, DeFi protocols, EVM dApps, or Web3 signing workflows in engagement scope.
Guides cloud penetration testing for AWS, Azure, and GCP enumeration, credential abuse, metadata SSRF, storage misconfigurations, IAM privilege escalation, and cloud lateral movement from web or internal footholds.
Guides OS command injection testing with shell metacharacter probes, blind exfiltration, filter bypass, and OS-specific syntax. Use when inputs reach shell commands such as ping, nslookup, file conversion, or when shell metacharacters alter application behavior.
Guides container and DevOps penetration testing for Docker escape, Kubernetes abuse, CI/CD pipeline secrets, package manager poisoning, and secrets enumeration from footholds or exposed services.
Guides database penetration testing for MSSQL, MySQL, PostgreSQL, MongoDB, Redis, and Elasticsearch. Use when database ports are open, SQL injection yields DB access, or linked-server abuse is suspected.
Guides insecure deserialization testing with format identification, language-specific gadget chains, and ysoserial/phpggc tooling. Use when Java, .NET, PHP, Python, or Ruby serialized objects appear in cookies, headers, APIs, or base64-encoded parameters including ViewState.
Guides digital forensics methodology for disk images, memory dumps, pcaps, and artifact analysis. Use when analyzing evidence to support incident response, malware triage, or pentest finding validation.
Guides GraphQL injection and authorization testing with introspection, field enumeration, batch attacks, and nested query abuse. Use when GraphQL endpoints, introspection, or query syntax are discovered.
Guides network and service penetration testing after port discovery. Provides port-to-skill routing, per-service quick reference, and general enumeration methodology for services with and without dedicated harness skills.
Guides hardware and physical access penetration testing with UART/JTAG/SPI analysis, firmware extraction, and logic analyzer techniques. Use when engagement ROE includes firmware extraction, debug interface access, or physical device compromise.
Guides Insecure Direct Object Reference testing with ID manipulation, parameter pollution, and horizontal/vertical privilege escalation techniques. Use when predictable IDs in URLs, sequential numbers, UUIDs, or authorization bypass indicators appear.
Guides initial access techniques including external attack surface mapping, credential abuse, phishing delivery chains, and client-side payload planning. Use during external engagement phases before exploitation.