"Kit" matching MCP servers:

A production-minded MCP server kit that provides OAuth 2.1/PKCE authentication, rate limiting, scope-gated tools, and two-phase confirmation for building secure MCP servers with custom tools, identity, and storage.

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 AI agents to safely operate on PostgreSQL data through typed MCP tools with per-identity RBAC, guardrails, and immutable audit logging.

A production-ready MCP server template with OAuth 2.1, RBAC, and audit logging for building secure, observable tool servers.

Provides an OAuth 2.1 authorization and login layer plus tool registration for MCP connectors, so a connector can be added to Claude clients with only a URL and password. It also shapes tool names, descriptions and input schemas for what Claude clients actually display, returning compact results and readable error messages.

Enables MCP clients to securely connect to your data through a remote deployment with OAuth 2.1 and PKCE, configurable sign-in, read/write consent, and audit logging. It exposes public tools without sign-in and authenticated tools that respect each user's permissions while validating arguments and recording every call.

Enables developers to build and run authenticated FastMCP servers with worker-thread execution for blocking tools, liveness probes, collision-safe file staging, and optional PDF/OCR helpers.

A local-first FastAPI and MCP safety kit for turning internal HTTP APIs into controlled MCP tools, with per-tool scopes, quotas, audit events, and default-deny web-access boundaries.

A local-first MCP attack kit for authorized security research and red teaming, demonstrating exfiltration, agent manipulation, and other abuse patterns via pluggable cases.

MCP server for Infinihash KYT providing real-time blockchain transaction monitoring, sanctions screening, and SAR generation. It enables MCP-compatible agents to screen wallets against OFAC SDN and generate compliance cases.

Enables AI assistants to inspect employee access, list failed onboarding events, and retry provisioning operations.

Enables MCP clients to read and, when explicitly enabled, modify Hudu IT documentation—companies, articles, assets, asset layouts, and password resources—through curated read tools, bounded paginated list operations, and policy-gated write/delete dispatch over the node-hudu SDK. Supports stdio and HTTP transports with per-caller credentials, read-only defaults, secret-read restrictions, resource-level search scoping, and redaction of credential fields.

Tenant-bound MCP server for agent spend authorization, policy guardrails, evidence, escrow settlement, and portable receipts. Works with Cursor and Claude Code via stdio; npm (@paybond/kit) and Python (paybond-kit).

Enables AI coding agents to reduce context window bloat by filtering noisy Git diffs, summarizing structural AST changes, simulating merges, and masking high-entropy secrets before they reach the model.

Open-source sample MCP server built to be audited: ships a paid-tool x402 demo, a deliberately scannable surface, and documented security posture. From Payload, makers of the MCP Launch Readiness Audit (48-rule scanner with a fix for every finding, hardening templates, stress-test harness, readiness verifier, CI workflow, branded audit report) and the MCP Monetization Kit.

Provides tools for autonomous CVE detection, enrichment, and remediation across managed repositories using GHSA and NVD data. It enables automated triage and pull request creation for dependency fixes based on configurable severity policies.

MCP server for securely managing GitHub & GitLab credentials in an encrypted vault, allowing AI agents to retrieve tokens by org or group name automatically.

Enables LLMs to query Kite Passport user info, sessions, wallet balance, and backend health via read-only MCP tools.

Enables agents to audit their own shipped artifacts for metadata and reverse-engineering exposures, apply equal-length overwrite mitigations, verify those mitigations hold, and generate truthful license and no-reverse-engineering notices. Also produces a probe kit for runtime observation resistance without modifying the artifact.

Enables MCP-capable runtimes to read agent message rooms, sign and post public messages, and create or verify Ed25519 contribution proofs for Technocore.