"Toll" matching MCP servers:

Unified MCP server for Plurity's Audit (GEO readiness), Toll (agent traffic & llms.txt), and Intelligence (question & topic monitoring) services, requiring a single API key.

Enables centralized proxy and orchestration of MCP servers, providing unified tool discovery, authentication, and secure request routing. It adds a human-in-the-loop approval gate for high-risk actions and audit logging.

MCP server that routes natural language requests to structured tool calls using a LoRA-tuned small language model, with built-in validation, retry, and fallback recovery.

Enables agents to send only the handful of tool definitions relevant to each question, ranking a large catalog by keywords and optional embeddings to cut context cost and wrong-tool picks. It can also run in front of any mcpServers config as a proxy that exposes just search_tools and call_tool instead of the whole catalog.

Routes external MCP tools behind a small on-demand discovery interface for Codex agents, using Jev to select relevant tools and providing fallback to the full inventory when confidence is low.

Enables AI agents to use paid MCP tools within a human-approved budget by holding a PayPal sandbox authorization, checking each call against policy, recording usage in a durable ledger, and capturing or voiding the authorization at session close.

Enables AI agents to securely discover, invoke, and manage tools through a hardened MCP endpoint with protections like injection detection, circuit breakers, retry backoff, response caching, context-window limiting, and state snapshots.

Enables AI agents to propose changes, obtain human or policy approval, and safely apply them with drift checks, idempotency, verified completion, compensation, and a tamper-evident ledger.

Enables role-scoped access to MCP tools through a policy-generated gateway that enforces content checks, session taint, budgets, pinning, and audit logging, blocking private-to-public data flows after untrusted input.

Enables AI agents to discover and execute tools via a secure MCP server with JWT authentication, RBAC, rate limiting, and audit logging.

Pay-per-call tools for AI agents via x402 (USDC on Base): web-to-Markdown, PDF, provenance (human vs synthetic), trusted lookup with citations, a 3.9M-business global directory, LatAm ID validators, agent-forged tools and 2,000+ proxied x402 services. No API keys, charged only on success, free trial without a wallet. Remote: https://toll402.dev/mcp · Local: npx -y toll402-mcp

A generic MCP proxy that filters which tools are exposed from a remote MCP server, reducing context window token usage by only loading the tools you actually need.

Reference implementation of a Model Context Protocol server for integrating external APIs as agent tools. Includes built-in tools for weather, database queries, web search, and calculations, with authentication, rate limiting, and logging.

Enables fault-tolerant agent tool execution with adaptive exponential backoff, full jitter, rate-limit 429 backpressure handling, and circuit breaker patterns to make AI agent calls resilient to transient failures. Plugs directly into MCP-compliant clients via JSON-RPC 2.0 stdio with zero external dependencies.

This MCP server wraps agent tool calls with fault-tolerant retry logic, including adaptive exponential backoff, full jitter, 429 rate-limit backpressure handling, and circuit breaker patterns. It enables autonomous agents to execute tools reliably without frontier LLM round-trips, using a zero-dependency Python stdio JSON-RPC interface.

Enables agents and MCP-compliant clients to resolve tool execution graphs as topological dependency DAGs, batching independent calls in parallel and speculatively prefetching arguments to cut latency and token burn. Runs deterministically as a zero-dependency JSON-RPC 2.0 stdio server for frameworks like LangGraph, CrewAI, AutoGen, and IDEs such as Cursor, Windsurf, and Claude Desktop.

A framework for building hot-loadable MCP tools that feature conversational negotiation, graduated permissions, and zero-downtime updates. It enables persistent state continuity across server restarts and provides a hypothetical evaluation system for tool actions.

Enables Claude Code to dispatch opencode CLI tasks as background processes with immediate task handle return, status polling, and result retrieval, avoiding tmux and log parsing issues.

An MCP proxy that embeds tool definitions and returns only the top-k semantically relevant tools per query, reducing context usage and routing calls to upstream MCP servers.

MCP router that fronts your other MCP servers behind a few small generic tools, so their tool definitions stay out of the context window and are discovered on demand.

Central registry for AI tools with vector search and knowledge graph, enabling intelligent tool discovery, chaining, and execution via MCP.

Graph-based tool retrieval for LLM agents. Builds a tool graph from OpenAPI/MCP specs and retrieves multi-step workflows via hybrid search (BM25 + graph traversal + embedding), recovering accuracy from 12% to 82% with 79% fewer tokens. Also works as an MCP Proxy to aggregate multiple servers behind 3 meta-tools.

Self-hosted daemon connecting Claude Code, Codex, and other coding agents across machines via MCP + A2A: delegate tasks, ask questions, share results, and wake sessions.

A single MCP server that provides dynamic access to thousands of APIs through discovery and proxy, eliminating the need to install and configure multiple separate MCP servers.

Enables coding agents to register tasks, report progress, save checkpoints, and report interruptions so an independent watchdog can automatically resume interrupted work from the same session or a saved checkpoint.

An MCP server that provides a long-running tool with configurable timeouts and periodic progress notifications, preventing Claude Code from disconnecting during healthy long tasks.

A proxy that reduces context token waste by exposing meta-tools (search_tools, load_tool, unload_tool) so the agent loads only needed tool schemas instead of all upstream tools.

A modular server implementation for Claude AI assistants with integrated tools, enabling Claude to perform actions and access external resources like file systems, web searches, browser automation, financial data, and document generation.

Lets AI agents drive a headless Chromium browser to open URLs, click, type, select options, read text/HTML, evaluate JavaScript, wait for elements, and capture real PNG screenshots. A persistent single session keeps cookies and SSO state across tool calls, and an optional live view mirrors pages in VS Code's integrated browser tab.

A standalone MCP gateway/proxy server that lets any MCP-compatible client search and call tools across multiple downstream MCP servers through a single mcp proxy tool.