MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
人看网页看板、AI 读纯文本的工单系统(MCP server)|Kanban + ticket system for humans and AI agents: plain-text API (Accept: text/plain) instead of JSON, 10 MCP tools, one Node process, one SQLite file, no Docker, no build step.
MIT -
Countbean plugin for Claude Code — hosted plain-text (Beancount) accounting. Published from the countbean monorepo; do not edit here.
MIT ★ 24 -
Reliable plain-text editing for agent systems. A sibling MCP to en-quire: literal range + anchor ops, no structural interpretation.
npm MIT -
countbean-mcpgithub.com/cputester5465/countbean-pluginupdated this year
MCP server for Beancount plain-text accounting: double-entry transactions written by an AI agent, validated with bean-check and committed to git.
pypi MIT -
A local-first, plain-text activity tracker with a pluggable schema and an MCP server — keep a structured record of your work for reviews, portfolios, and credentialing.
MIT ★ 2 -
MCP server that lets AI agents read Excel, ODS, CSV, JSON, SQLite, Parquet and MTSV files, or a folder, as plain text
MIT ★ 1 -
news-tokgithub.com/dongquoctien/news-tokupdated this year
Turn articles, plain text, or URLs into short TikTok/Reels/Shorts videos. 100% local rendering, orchestrated by Claude Code CLI or the Web Studio.
MIT -
Turn STEP files into an LLM Context Pack: exact measurements, features and assembly facts in plain text.
pypi AGPL-3.0-only -
html-mcpgithub.com/MukundaKatta/html-mcpupdated this year
MCP server: strip HTML tags, extract plain text, and sanitize HTML.
MIT -
MCP server for Beancount: query, validate, and safely edit plain-text ledgers with an LLM
-
pdf-generator-mcpgithub.com/rumblingb/pdf-generator-mcpupdated this year
Generate PDF documents via MCP protocol. Convert HTML, plain text, and URLs to professional PDFs.
npm MIT -
Open source AI data analyst that answers questions about your data in plain English. Governed by a dbt semantic layer, runs locally on DuckDB, refuses to guess when it is unsure, and publishes execution accuracy on every commit. Text to SQL you can put in a board deck.
MIT ★ 1 -
prompt-capsulegithub.com/FeilixX/prompt-capsuleupdated this year
Short-lived text/plain URLs for long prompts and agent instructions
MIT -
Lightweight high-performance SSH MCP server in Go — single binary, context-friendly plain-text output, strict credential safety. 轻量高性能的 Go SSH MCP 服务器
MIT ★ 2 -
Extract numeric values from config files, data files and plain text.
npm MIT -
Extract string values from config files, data files and plain text.
npm MIT -
Yet Another Text-based TODO App — every task is a plain markdown file you own
MIT -
A design document format built for the age of AI agents. Zenith is a plain-text format and engine for design files — posters, decks, books, social graphics, and more
Apache-2.0 ★ 23 -
Find hidden prompt-injection text: ASCII smuggling, zero-width/emoji stego, bidi, homoglyphs. One layer, not a full defense: it does not judge prompt injection written in plain, visible words.
MIT -
@nmbrai/claude-codeactive
nmbr for Claude Code — a channel plugin (text your Claude Code session from your phone, approve permissions on nmbr cards) that is also a plain MCP server for any MCP host.
npm MIT -
MCP server and CLI that turn one sentence into a finished deck, doc, or dashboard — plain-text source, self-contained web page out. Runs on your own API key or fully local with Ollama.
MIT -
PrivacyScrubber PII Masking MCPno update data
Zero-Trust Data Sanitization (ZTDS) MCP Server for Claude Desktop, Cursor, Windsurf, and other AI clients. Automatically intercepts and redacts PII, credentials, and sensitive corporate data locally on your machine before they are sent to external LLMs. ### 🔒 100% Local & Airplane Mode Verified All sanitization runs inside local RAM. No data ever leaves your computer, ensuring compliance with GDPR, HIPAA, SOC 2, and internal GRC policies. Zero network calls, zero server logs. ### 🛠️ Supported Formats & File Types Locally parse and sanitize: - **Documents:** `.docx` (Microsoft Word) parsed locally with Zero-Server extraction. - **Code & Configs:** `.js`, `.py`, `.go`, `.rs`, `.json`, `.yaml`, `.ini`, `.env`, and source code files. - **Data & Logs:** `.txt`, `.log`, `.csv`, `.tsv`, `.md`, and plain text outputs. ### 🏷️ Redacted Entity Types - **Personal Data:** Names (`NAME`), Emails (`EMAIL`), Phone Numbers (`PHONE`). - **Credentials:** API Keys, passwords, tokens, SSH keys, database URIs. - **Identifiers (`ID`):** SSNs, Tax IDs, credit card numbers, and custom regex patterns. ### 🎛️ 22+ Specialized Industry Detection Profiles Specify detection rules tailored to your workload (requires PRO key): - **Engineering & IT:** `Dev`, `Tech`, `Security`, `Agents` - **Corporate & Operations:** `General`, `Bizops`, `Sales`, `Support`, `Marketing`, `Creative` - **Financial & Professional:** `Finance`, `WealthMgmt`, `Insurance`, `Accounting` - **Legal & Compliance:** `Legal`, `Compliance`, `CCPA`, `HR` - **Healthcare & Research:** `Medical`, `Pharma`, `Academic` - **Personal Use:** `Personal` ### 🔧 Available Tools - `sanitize_text` — Scrub PII/Secrets locally in text context. - `reveal_text` — Reversible token mapping (Original -> Token -> Original). - `sanitize_file` — Local sanitization of raw plain text and DOCX files.
smithery -
Claude Code plugin that turns conceptual Markdown or plain text into source-anchored knowledge graphs. Build, ground, query, visualize, and challenge relationships with provenance, retained failures, and clearly separated unverified hypotheses.
MIT -
Convert Markdown to Slack, Microsoft Teams, Jira, and plain text with an MCP server. Supports CommonMark and GitHub-Flavored Markdown (GFM) and provides platform-specific formatting for AI agents, LLMs, and automation workflows.
npm MIT -
The personal operating system for AI agents: memory, tools and hooks in one relational data bank of plain text, for Claude Code, Codex and Cursor. Memory is space; time is CPU cycles.
Apache-2.0 -
Write your text in the code, ship it in every language: an i18n compiler and a ~3 KB runtime for React, Vue, Svelte, Next.js, Nuxt, SvelteKit, Astro and plain HTML. Typed params, no key files, and a CI gate for broken translations.
npm MIT ★ 4 -
Free, open source, git-native API client. A local-first Postman, Insomnia and Bruno alternative for macOS, Windows and Linux: plain-text .tiger collections, works offline, no account, built-in MCP server for Claude and Cursor. MIT.
MIT ★ 107 -
Self-hosted MCP server that makes your Obsidian/markdown vault searchable — text, PDFs, Office docs, images, audio — from any MCP client. Hybrid retrieval over a typed, governed corpus, sub-second at 50k notes; your files stay plain markdown.
pypi AGPL-3.0 ★ 9 -
The Universal AI Video Engine: Autonomous 3D Motion Graphics, VFX/CGI, Higgsfield Cinema DoP & Remotion NLE for Claude Opus, GPT, Gemini & AI Agents. Turn plain text into Hollywood-grade cinema via MCP & CLI.
Apache-2.0 ★ 3 -
Free, open-source MCP server (Apache-2.0) giving Claude plain-English access to official Austin & Travis County data — property records, permits, flood zones, city code text, schools, plus active MLS listings. Built by Neuhaus Realty Group.
Apache-2.0 ★ 6 -
Own your AI memory, don't rent it. A native Claude Code skill AND a standalone, zero-dependency MCP server over one local SQLite file. git clone and go — full-text knowledge graph, plain-markdown, you own the data.
MIT ★ 7 -
Enterprise-grade MCP server for live docs, best practices, code audit. Context7 alternative — 793+ libraries, evidence-verified answers (never generic), smart dispatch (use gt mcp), plain-text intent routing, telemetry, SSRF-hardened multi-source fetcher,
npm Elastic-2.0 -
quickbooks-mcpgithub.com/nichewizard/quickbooks-mcpupdated this year
A QuickBooks MCP server built for real books. Exposes QuickBooks Online as callable tools over the Model Context Protocol, with every create, update and delete gated behind human approval and a plain-English summary, and all QuickBooks text treated as untrusted input rather than instructions.
Apache-2.0 ★ 5 -
KG-SSOT: verifiable knowledge graphs in plain text. A formal contract for .kg.yaml and .kg.json (JSON Schema) with a conformance suite, reference readers (Python, JavaScript), CI gate, store, query, diff and MCP server. Release mirror, Apache-2.0.
npm Apache-2.0 -
Self-hosted paper manager for researchers and labs: read and annotate PDFs in the browser, keep them as plain files, full-text and semantic search, AI chat that cites the page, an MCP server, a browser extension and Zotero import. One docker compose.
MIT -
scrcpy-mcpgithub.com/JuanCF/scrcpy-mcpno update data
scrcpy-mcp gives AI agents real vision and control over Android devices. It exposes screenshots, taps, swipes, text input, key events, UI inspection, and arbitrary ADB shell commands as MCP tools, so an agent can drive a phone or tablet the way a person would. When a scrcpy session is active, input and screen capture run 10–50× faster than plain ADB, making real-time automation, on-device testing, and UI agent workflows practical. Highlights - Fast screenshots and live screen capture - Tap, swipe, drag, long-press, scroll, and text input - UI hierarchy dump and element search by text, resource-id, or class - Arbitrary ADB shell command execution - Optional scrcpy session for low-latency control Works with any ADB-reachable Android device (USB or Wi-Fi). Configure ADB_PATH, ANDROID_SERIAL, and scrcpy paths as needed.
smithery -
Open plain-text file format for AI memory. Your assistant's long-term memory as .dai files on your disk: readable by Claude, GPT, Gemini, Cursor, local models and grep (all LLM models work). MCP server + hooks for Claude Code, Claude Desktop, Cursor, Windsurf, Codex. 83% LongMemEval-S (GPT-4o), 92% (Claude Fable 5), 10x fewer tokens.
Apache-2.0 ★ 55 -
**Persistent memory for AI agents that you can actually read.** Kybase stores what your agent learns as plain Markdown notes with `[[wikilinks]]`, searched by keyword and by meaning. It runs locally: notes live in an embedded Postgres on your machine (`~/.kybase`), with no Docker, no server and no account. **What your agent gets (18 tools)** - **Search** — `search_notes`: hybrid keyword + semantic search that returns the section each excerpt came from - **Read** — `get_note` (whole, windowed or one section), `list_notes`, `list_tags`, `list_folders`, `indexing_status` - **Write safely** — `create_note`, `append_to_note`, `replace_in_note`, `update_note` with conflict protection, `delete_note` / `restore_note` (trash) - **Link** — `get_backlinks`, `get_neighbors`, `get_graph` over the wikilink graph - **Organize** — `create_folder`, `update_folder`, `delete_folder` **Setup** Keyword search works right away. For search by meaning, pick an embedding provider in the settings: local **Ollama** (`ollama pull embeddinggemma`) keeps everything on your machine; **Google** or **OpenAI** need an API key and send note text to that provider. **Your data stays yours** `kybase-mcp export vault.zip` writes every note out as plain Markdown, folders as directories, readable in any editor, Obsidian included. **Want a UI?** The full Kybase app adds a web editor, a knowledge-graph view, share links, and one memory shared by Claude Code, claude.ai and Cursor: [github.com/Kyrzin/kybase](https://github.com/Kyrzin/kybase) Open source, AGPL-3.0.
smithery -
ClarityOSgithub.com/RishiBuilds/ClarityOSupdated this year
an agentic tool that simplifies complex institutional texts to meet readability targets using a multi-agent feedback loop.
MIT ★ 1 -
Read the Subnotes notebooks you connect and add tasks to them — so "put that on the backlog" lands in your notebook, not the transcript. Needs the Subnotes app, macOS only.
MIT