Security MCP servers for scanning, threat intelligence and audits: let an AI assistant check code, dependencies and infrastructure for risks. 3,537 servers in this category.
Verifies that streaming redactors and guardrails do not leak secrets when input is split across chunks by generating automated pytest regression tests.
$claude mcp add chunk-invariance -- uvx chunk-invariance
Acts as a governance proxy between AI agents and MCP servers to enforce allowlists, dry runs, human confirmation, and audit trails.
$claude mcp add mcp-airlock -- uvx mcp-airlock
Enables AI agents to manage Tor circuits, rotate exit nodes, and execute anonymized web requests through a built-in MCP server.
$claude mcp add torproxy -- docker run -i --rm docker.io/hapheus/torproxy:2.7.7
Enforces code-of-conduct policies across MCP connectors using per-user memory to prevent sensitive data combinations and prompt injection.
$claude mcp add aggrete-7f66c8 -- uvx aggreteEnforces a code-of-conduct policy across MCP connectors, refusing forbidden tool combinations before the upstream call runs.
$claude mcp add aggrete -- uvx aggrete
Provides a tamper-evident ledger for agent actions using hash-chaining to ensure logs cannot be quietly rewritten or altered.
$claude mcp add arcaeon -- uvx arcaeon
Allows an AI client to consult ENS security measures, calculate applicability matrices, and generate auditable compliance checklists.
$claude mcp add ensmcp -- uvx ensmcp
Lets AI agents autonomously run 150+ security tools for pentesting, vulnerability discovery, and bug bounty automation.
Records AI trading decisions in a tamper-evident hash chain and recalls past trades before your agent places an order.
$claude mcp add tradememory-protocol -- uvx tradememory-protocol
Enforces AI accuracy by using deterministic tools to verify claims against ground truth in binaries and source code.
Lets AI clients investigate threats, run SOAR actions, query Google Threat Intelligence, and audit cloud security findings in Google SecOps.
Enables AI clients to perform defensive SOC automation, threat intelligence lookups, and MITRE-driven incident investigation using Wazuh SIEM.
Provides agentic memory for cyber threat intelligence by extracting entities, building STIX 2.1 knowledge graphs, and performing intent-aware retrieval.
$claude mcp add zettelforge -- uvx zettelforge
Enables AI clients to perform reverse engineering on ELF, Mach-O, and PE binaries using Ghidra to decompile, disassemble, and trace value flows.
$claude mcp add pyghidra-lite -- uvx pyghidra-lite
Lets an AI client list, inspect, and grant or deny pending CTRLRun approval requests.
$claude mcp add ctrlrun -- uvx ctrlrun
Lets an AI client run allowlisted, audited commands and structured file edits on Linux, macOS, and Windows hosts.
$claude mcp add --transport http sentinelx-cloud-core https://mcp.sentinelx.app/mcp/mcp
Answers offline CVE, exploit, and vulnerability questions from a local SQLite index built from public security feeds.
$claude mcp add crowbar -- uvx crowbar-cli
Enables AI assistants to query the Vulners vulnerability database for CVE information, security bulletins, and software package auditing.
$claude mcp add vulners-mcp -- uvx vulners-mcp
Provides a safe SQL gateway for AI agents with SELECT-only validation, access policies, data masking, and an evaluation harness.
$claude mcp add querypilot -- uvx querypilot
Lets an AI client search, triage, and import security findings into DefectDojo via its REST API v2.
$claude mcp add defectdojo-mcp -- uvx defectdojo-mcp-server