"Desktop software without licensing requirements" matching MCP servers:

Provider-neutral MCP server for managing VirtualBox, VMware Fusion, and VMware Workstation, offering tools for VM lifecycle, configuration, snapshots, guest operations, networking, and artifact resolution.

Enables AI assistants to read, search, organize, and draft email, manage calendar items, and look up contacts through the user's already signed-in classic Outlook desktop app on Windows, with outgoing actions defaulting to drafts for review.

Connect your Umami Analytics to any MCP client to derive insights from natural language.

Enforces structured, evidence-guided software engineering tasks with cognitive actions (investigate, plan, verify, remember) and persistent state for LLM-based coding agents.

Enables governed multi-agent software factory operations through eight typed, audited tools for Sprint planning, submission, decisions, inspection, cancellation, and resumption.

self-hosted development harness (Go engine + CLI + desktop, Linux first) · brief → requirements → spec → plan → build → review → release · verdicts are exit codes, never model opinions · local models first (llama.cpp, vLLM) beside any OpenAI-compatible or Anthropic endpoint · operable by humans or by other agents over MCP with recorded, attributed decisions

Easily build MCP tools for Claude desktop app.

Haft is an MCP server that enforces engineering discipline for AI-assisted software delivery, providing tools for problem framing, decision contracts, and commission-based execution harnesses.

An MCP server for querying Grafana Loki directly with a discovery-first workflow — labels, values, series, and LogQL queries without requiring Grafana.

An MCP server providing web search, image search, and page scraping tools to LLMs without requiring API keys.

Self-hostable remote terminal assistance over MCP. Agents create one-command connection invitations, execute commands, and manage interactive sessions on user-authorized macOS, Linux, and Windows computers without requiring an SSH daemon.

MCP server that parses 3GPP .docx specifications into structured sections with full-text search, letting LLMs browse, search, and follow cross-references without a vector database.

MCP server — read-only log access for AI assistants. Debug your Linux server with AI, without giving the AI shell access.

Lets a model use hundreds or thousands of MCP tools without loading them: it puts five tools in front of a large MCP server or gateway, so the model searches for the tool it needs and calls it. Search combines BM25 with a multilingual embedding model that runs inside the proxy.

An MCP server that gives AI agents structured read/write access to a story-based project backlog. Agents can list stories, read content, update status, and append notes — all backed by plain markdown files that live inside your project repository. There is no shared server. The backlog files live in your repo under requirements/, committed and versioned alongside your code

Manages an agent's long-term memories as versioned markdown files with searchable SQLite indexes, supporting lifecycle states, conflict-gated recall, revision, and explicit forgetting without automatic deletion.

Local-first shared memory and task coordination for AI coding agents. One Go binary, MCP server, markdown files you own. Hooks for Claude Code and Codex CLI (and their desktop apps).

A local AST and type-aware Go repository context indexer that uses CHA to map symbols, call graphs, dependencies, and architecture patterns. Built for AI coding agents that need to navigate and reason about large Go codebases without reading raw files.

Local, open-source multi-account MCP broker — one endpoint, every account. Add credentials once and switch identity without reconnecting, from any LLM client.

Manage your entire notification infrastructure using natural language. Trigger workflows, create users, manage preferences, update tenant branding, and access docs — all from Cursor, Claude Desktop, or Windsurf. 24 tools covering email, SMS, push, WhatsApp, Slack, MS Teams, and in-app notifications.

A Zotero MCP server built for agent safety: a capability-tagged tool registry (every command carries operation/requires/destructive annotations), preview-first writes behind a mutation journal, and read-only defaults. Single Go binary with an MCPB one-click manifest for Claude Desktop.

Transparent Go proxy that intercepts, signs, rate-limits, redacts, and audits all MCP JSON-RPC tool calls without modifying client or server. Stores to JSONL or SQLite with HMAC-SHA256 signatures.

Mishmar is a risk management layer for trading bots and AI agents: every order is checked against your limits before it reaches the exchange, large orders wait for your approval, and no key can trade alone. Its MCP server lets a model propose orders on Hyperliquid and Binance and read positions, PnL and approvals, without ever seeing a key.

Give Claude Code, Cursor, and other AI agents safe access to your real infrastructure — without giving them raw SSH access.

Enables AI agents to list, describe, and invoke scoped capabilities so they can perform brokered HTTP or SSH operations without ever receiving the underlying credential. All uses are deny-by-default, revocable, expiring, and recorded in a hash-chained audit log.

SuperDev is a local-first desktop app + MCP server that gives AI coding agents (Claude Code, Codex, Cursor — anything MCP-compatible) access to the runtime: live and historical logs across services, breakpoints in already-running processes (same PID, no restart), real browser control, and DAG deploy pipelines — with human approval and an audit trail on protected writes.

Zero trust gateway for MCP servers. Aggregate, filter, and securely access MCP tools from anywhere without VPNs, open ports, or exposed endpoints. Built on OpenZiti, zrok, and Agora with cryptographic identity, mTLS, per-client isolation, and tool-level permission control.

An MCP server that provides Claude Code with 99 methodology skills, seed patterns, decision logging, and memory tools to improve software development decisions. It acts as a cognitive scaffold layer, enabling better working methods rather than just additional data.

Enables AI agents to publish themselves, discover each other, and establish authenticated encrypted connections without central infrastructure, using a decentralized agent-to-agent networking protocol.

High-performance Model Context Protocol (MCP) proxy, multiplexer, and dynamic JSON schema pruner in Go. Compresses verbose tool schemas across 4 optimization tiers, cutting LLM context window token consumption by up to 91.4% for Claude Desktop, Cursor, and Windsurf.

Enables AI editors to install, monitor, and query Stockyard's collection of 150 self-hosted developer tools. Manages lightweight Go-based tools with embedded SQLite databases without requiring Docker or external dependencies.

Enables finding recurring charges, subscriptions, price increases, and possible double charges from a bank or card statement CSV, along with yearly cost estimates, without any bank login or uploads. It also inspects a CSV to show detected columns and row counts before analysis.

A self-hosted MCP server that provides capability packs for AI agents to perform browser, desktop, git, and code actions via schema-validated JSON calls, enabling reliable agentic workflows even with small open-weight models.

Control-plane proxy that sits in front of MCP servers, enforcing per-identity policy (YAML/OPA/Cedar), budget limits, and an audit trail on every call. Includes real-time anomaly detection that auto-blocks a compromised agent without a human in the loop.

An MCP server for semantic code search & navigation that helps AI agents work efficiently without burning through costly tokens. Instead of reading entire files, agents can search conceptually and jump directly to the specific functions, classes, and code chunks they need.

Enables MCP clients such as Cursor and Claude Desktop to query a GRC platform over stdio or HTTP, with read-only access by default and opt-in create, update, and delete tools subject to platform authorization.

Enables AI coding assistants to inspect OpenAPI specs, fuzz live REST endpoints with boundary and adversarial vectors, replay failing anomalies deterministically without AI tokens, and scaffold OpenAPI specs from source routes.

Enables sending emails via Amazon SES, SMS/MMS/RCS messages through AWS End User Messaging, and sharing files via CloudFront-signed download links, all from AI assistants like Claude Code or Claude Desktop.

Zero-dependency, RAM-only HTTP diagnostic proxy and MCP server in pure Go. It intercepts traffic in RAM and replaces secrets with salted SHA-256 correlation tags so AI agents can debug auth flows without credentials entering LLM contexts or touching disk.

Enables agents to reduce tool noise by exposing only two meta-tools for searching downstream MCP servers and calling their tools on demand. Connections are lazy and tool lists are cached, so configured servers can be used without loading every tool upfront.