"Information about guardrails" matching MCP servers:

MCP server for AI agent security guardrails. Provides input validation, prompt injection detection, PII redaction, output filtering, policy enforcement, rate limiting, and comprehensive audit logging.

Exposes LLM output validation (schema compliance and prompt-injection detection) as an MCP tool, enabling any MCP-compatible agent to apply real-time guardrails without importing the codebase.

Provides AI security guardrails through Javelin's platform to detect harmful content, prompt injection attempts, and language policies. Enables comprehensive content safety analysis with trust & safety detection, prompt injection protection, and language identification.

Enables AI agents to run read-only AWS security audits and receive normalized, severity-ranked findings about account posture, such as stale access keys, missing MFA, public buckets, open security groups, and CloudTrail coverage.

A robust Model Control Protocol server that enables AI agents to access real-time cyber threat intelligence and detailed information about vulnerabilities, threat actors, malware, and other cyber-security entities.

A read-only MCP server for the Qualys PCI Merchant API that lets LLM assistants answer questions about PCI compliance posture, such as which hosts are failing PCI or listing high findings.

Agentic KYC/AML compliance server with tools for sanctions screening, identity verification, and risk assessment, where AI orchestrates discretionary checks within deterministic compliance guardrails.

An MCP server that exposes macOS-native device trust signals — the facts about a Mac that cannot be gathered from a Linux container or a cloud runner. All tools are strictly read-only.

An MCP server that exposes 43 domain tools behind 4 meta-tools, applying guardrails such as budgets, rate limits, schema validation, human-in-the-loop confirmation for mutations, PII redaction, and an append-only audit log while reducing context window usage.

This server gives AI assistants controlled read and write access to a Firefly III personal finance instance over MCP, exposing 139 operations across 24 entities with optional read-only mode and support for both stdio and authenticated HTTP transport.

A Model Context Protocol (MCP) server for Socket integration, allowing AI assistants to efficiently check dependency vulnerability scores and security information.

A prompt-injection-aware MCP server demonstrating guardrails for agentic tool use, including permission tiers, untrusted-content quarantine, and heuristic injection detection, with a fictional ticket-triage assistant.

Enables an AI agent to create and capture orders, refund captures, and pay suppliers through PayPal while the operator's policy fixes the payee, currency, and credential behind a broker that the agent never sees. Adds spending limits, human-approval thresholds, server-derived idempotency, and a hash-chained audit ledger, so a compromised or injected agent can only pick an operation and bounded values.

Provides access to ODEI's constitutional knowledge graph, AI safety guardrails, and EVM smart contract auditing tools. It enables users to query structured domain nodes, validate agent actions, and perform security audits directly through an MCP client.

Exposes Microsoft 365 privacy/compliance audit checks as MCP tools, enabling natural language queries about tenant configuration, orphaned documents, Conditional Access policies, DLP policies, and Copilot usage.

Enables Claude Desktop and Claude Code to inspect a Postgres schema and run read-only SQL queries against a demo database through the tools describe_schema and query_readonly. Every proposed query passes the same deterministic guardrails — SELECT-only, allow-listed tables, enforced row limits, statement timeouts and a read-only role — before execution.

Sovereign AI Security Suite & Threat Scanner. Real-time detection for phishing URLs, zero-day web threats, quishing QR codes, prompt injection defense, carrier disarming (AI-CDR), and autonomous agent safety guardrails.

MISP (Malware Information Sharing Platform) MCP server with built-in prompt injection defense via prompt-defense-audit

Compliance and guardrails infrastructure for AI agents, enabling safe operations within regulatory boundaries like GDPR and EU AI Act.

Secure MCP server for safe, read-only DB access by AI agents, with SQL guardrails, table allowlists, PII masking, and audit logs

Enables natural-language queries about Azure resource compliance, including VM compliance, patch status, orphaned RBAC, and infrastructure health, through read-only MCP tools.

Enables AI agents to securely read and write SAP ABAP development objects via the ADT REST API, enforcing policy-based guardrails such as package/namespace allow-lists, read-only mode, human approvals, and audit logging.

Enables AI assistants to execute malware analysis tools on a REMnux system via Docker, SSH, or local connections. It provides automated file-type analysis, structured tool discovery, and security guardrails for streamlined malware investigation.

MCP server for AI agent guardrails — validates and signs agent decisions (OAA tokens) so third parties can verify them without server access

Guardrails for AI coding agents that enforces rules via compliance receipts, blocking destructive actions and syncing rules across all AI tools.

AI payment guardrails for Python & MCP: spend-capped agent identity (KYA), dry-run/budget/approval gates, prompt-injection defense, encrypted secret vault.

Enables AI agents to inspect database schemas and run strictly read-only SQL queries with guardrails against destructive statements, multi-statement chaining, and unbounded scans. Supports SQLite natively and PostgreSQL optionally.

Enables AI assistants to safely explore and query MySQL, MariaDB, PostgreSQL, SQLite and StarRocks databases through read-only tools for schema listing, table description, schema search, row sampling, query execution, and plan explanation. Enforces least-privilege accounts, engine-level read-only sessions with timeouts, parsed SQL guardrails, and column masking so runaway or unsafe queries are refused.

Enables AI assistants to validate individual or batch email addresses for deliverability, and to check account plan information, through MCP tools.

Enables local detection of protected health information in source code before it reaches LLM prompts, log statements, or analytics calls.