MCP Server Marketplace
All MCP servers
24 servers on this page
Agent Harnesses
io.github.RyanAlberts
Lets an assistant query a curated catalog of AI agent harnesses to recommend, compare, and pick one for a given model and task.
Mcp
io.github.Runware
Lets an assistant run Runware's image, video, audio, 3D, and upscaling models and browse its model catalog.
Runeward
io.github.Runewardd
Runeward wraps AI agent tool calls in policy checks, approvals, sandboxed execution, budgets, and a signed audit trail.
Gatekeeper
io.github.Runestone-Labs
Routes an assistant's shell, file-write, and HTTP tool calls through a local Runestone Gatekeeper server for policy checks, approval, and audit.
Agentpay Sentinel Mcp
io.github.Rumblingb
Runs pre-flight security checks on AI agent payment requests, returning PASS or BLOCKED before a transaction executes.
Agentpay Gateway Mcp
io.github.Rumblingb
A self-hosted gateway that routes an assistant's calls to 42 backend MCP servers with per-call credit billing and one API key.
Agentpay Feed Mcp
io.github.Rumblingb
Lets an assistant poll a real-time newsfeed of MCP tool registrations, capability announcements, and market events, and optionally publish new entries.
Agent Wallet Mcp
io.github.Rumblingb
Lets an assistant keep a local sandboxed spend ledger for agents, with balances, monthly budget caps, transfers, and invoice records.
Agent Audit Mcp
io.github.Rumblingb
Gives an assistant a local, tamper-evident audit log of agent actions, with hash-chained events, search, history and chain verification.
Compendio Mcp
io.github.RuloGB
Indexes a project's Markdown documentation locally and lets an assistant search and read the right sections instead of grepping whole files.
Assay MCP Server
io.github.Rul1an
A local MCP proxy that enforces allow/deny policy on routed tool calls and records offline-verifiable evidence bundles for review.
Assay
io.github.Rul1an
Assay is a local policy gate and evidence recorder that allows or denies MCP tool calls before they run and produces offline-verifiable audit bundles.
MySQL Legacy MCP
io.github.Rufflet
Lets an assistant inspect and query legacy MySQL 5.0-5.6 databases, with read-only access by default and opt-in write tools.
Workspaceguard
io.github.RudrenduPaul
Wraps the WorkspaceGuard CLI as one generic tool so an assistant can run usage, quota, and workspace commands and read their JSON output.
Workloadtruth
io.github.RudrenduPaul
Classifies a running GPU workload as TRAINING, INFERENCE, or IDLE from GPU telemetry alone, and verifies a local hash-chained audit log.
Truthroute
io.github.RudrenduPaul
Sends one prompt to several LLM providers and returns a validated divergence score, exposed as an MCP tool over stdio.
Truesignal
io.github.RudrenduPaul
Lets an assistant run the truesignal CLI to pull OSINT and security feed items from five official sources, with provenance labels.
Toolgovern
io.github.RudrenduPaul
Wraps the toolgovern CLI as a single generic MCP tool so an assistant can validate agent-tool policy rules before tool calls execute.
Tokentrust
io.github.RudrenduPaul
Runs a context-reduction proxy against a labeled task corpus and reports measured token and cost savings versus the proxy's claimed numbers.
Tenantguard
io.github.RudrenduPaul
Runs TenantGuard's tenant-isolation audit of a self-hosted multi-tenant AI-agent deployment config and returns structured findings to an assistant.
Teamspend
io.github.RudrenduPaul
Compares AI coding tool spend before and after a migration, pulling real cost numbers from vendor admin APIs or local session logs.
Taskswarm
io.github.RudrenduPaul
Wraps the TaskSwarm CLI as one generic MCP tool so an assistant can report and read parallel coding-agent session status.
Swarmmesh
io.github.RudrenduPaul
Lets an assistant share namespaced context and keyword-searchable memory with other agent processes through a local SwarmMesh coordination server.
Swarm Rd Orchestrator
io.github.RudrenduPaul
Gives an assistant three tools to append findings to and read back a shared, durable event log for parallel research agents.