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.
-
CLI and MCP server that turns your GitHub PRs and JIRA tickets into a searchable markdown work log for performance reviews and resumes.
npm MIT -
The work log your team's AI agents write. A self-hosted web app plus an MCP server: daily updates drafted from git and the session, tasks, time measured from agent sessions, project context and client access. Works with Claude Code, Cursor, Codex, Gemini CLI, Claude and ChatGPT. Free for any team.
NOASSERTION -
console-mcpgithub.com/rohithgoud30/console-mcpupdated this year
MCP server for streaming macOS Console.app and iOS device logs — works with Claude, Copilot, and any MCP-compatible AI assistant
MIT ★ 1 -
Model Context Protocol (MCP) server for a mileage log for deductible driving: trips, effective-dated rates, summaries and a CSV export. Mileage log for deductible driving: trips, effective-dated rates, summaries and CSV export. Works with Claude Desktop, Claude Code and Cursor.
MIT -
CLI and MCP server that turns your GitHub PRs and JIRA tickets into a searchable markdown work log for performance reviews and resumes.
npm MIT ★ 6 -
Model Context Protocol (MCP) server for an equipment maintenance log and schedule: service entries, costs, and what is due or overdue. Equipment maintenance log: assets, service entries with costs, and a due report computed from stored dates. Works with Claude Desktop, Claude Code and Cursor.
MIT -
Change engine for the PerfoAds MCP server (works with Google Ads): propose, check with Google, approve, apply once, read back, log and undo.
npm Apache-2.0 -
Chain of Thoughts (zoth-cot): zero-dependency working memory and reasoning log for AI agents, with an MCP server. Free and open source.
-
@felipesauer/mnemagithub.com/felipesauer/mnemaupdated this year
A tamper-evident, local-first audit trail for AI-agent work — an MCP server giving Claude Code, Cursor and other agents typed tools behind workflow gates, with every action recorded in a SHA-256 hash-chained log.
npm MIT -
The open-source work log for human + AI teams — one timeline of human work and AI-agent sessions, turned into any report.
MIT ★ 4 -
MCP server for managing development logs and working notes
npm Apache-2.0 -
Terminal note-taking app for developers. CLI notes, work logs, daily journal, with SQLite FTS search and MCP AI integration.
pypi MIT License -
MCP server for Timely time tracking: question-shaped, read-only tools for projects, time spent and work logs
MIT -
Let Claude Code log into websites and use your API keys without ever seeing them. Works from your phone.
npm MIT -
Laravel-Debugbar-MCPgithub.com/sergey0002/Laravel-Debugbar-MCPupdated this year
MCP for working with laravel-debugbar logs
NOASSERTION ★ 1 -
caishino update data
Read Chinese platforms from any MCP client: Bilibili, Douyin, Xiaohongshu, WeChat articles, Zhihu, and any web page. Content is read through your own logged-in browser session, so pages that require a login work too.
smithery -
Log work to Resource Optimiser and Jira from the terminal, an MCP server for AI IDEs, and a terminal UI.
npm MIT -
MCP server for UEFI work: boot an .efi under OVMF and explain the outcome, decode EFI_STATUS and GUIDs, check PE images, parse OVMF logs and firmware volumes.
MIT -
Self-hosted time tracking API with MCP server and Claude Code plugin — log work hours by client, project and task
-
Local LLM does Claude's bulk work: reads logs and files, researches the web, writes test-gated code.
MIT ★ 2 -
Know your agent stopped working before your customer does. Alerts on missing activity, failure streaks and undelivered messages from a JSONL event log. CLI, library, and MCP server. Zero dependencies.
MIT -
Local LLM does Claude's bulk work: reads logs and files, researches the web, writes test-gated code.
npm MIT -
MCP server for iOS app testing — see inside your running app: network calls, WebSocket frames, logs, UserDefaults, screen layout, and API mocking. Works with Claude Code, Cursor, Codex, Windsurf.
npm MIT -
OSINT Agent Skills — 24 MCP tools for autonomous OSINT investigation. DNS, WHOIS, CT logs, Shodan, breach data, geolocation, crypto tracing. Works with Claude Code, Cursor, any MCP client.
npm MIT -
Local-first personal health record exposed over MCP. Labs, meds, diagnoses, wearables and food log in your own Postgres, with deterministic safety checks. Works with any MCP client.
AGPL-3.0 ★ 3 -
Know your agent stopped working before your customer does. Alerts on missing activity, failure streaks and undelivered messages from a JSONL event log. CLI, library, and MCP server. Zero dependencies.
npm MIT -
Keep ClickUp in sync from your terminal, git and AI agents (Claude Code, Cursor, MCP): plan the day, start/finish tasks, log every commit, see who works on what.
npm MIT -
MCP server for iOS app testing — see inside your running app: network calls, WebSocket frames, logs, storage, screen layout, and API mocking. Works with Claude Code, Cursor, Codex, Windsurf.
MIT ★ 1 -
Mitra is a modular, stateless Model Context Protocol (MCP) server that integrates Clockify, WakaTime, and Azure DevOps. It enables developers and local AI assistants to fetch active projects, manage Azure DevOps work items (cards), and log time entries directly to Clockify using a unified workflow.
★ 2 -
Curtis is an MCP server that runs your LinkedIn outreach from your own machine: it logs in once, works through your contact list, and sends connection requests slowly enough to look like you.
npm MIT -
Self-hosted email MCP server and IMAP/SMTP gateway for AI agents: one token per agent with its own mailboxes and scopes, an audit log, and a passkey webmail PWA. Works with Claude Code, Codex, Cursor, OpenClaw.
NOASSERTION ★ 1 -
MCP server that lets one coding agent hand heavy read-only work (scanning files, filtering logs, analyzing bundles) to another agent CLI running in the background, and read back only a compact report.
npm MIT -
truthgithub.com/blasrodri/truthupdated this year
Deterministic fact-checker for AI coding agents — verify an agent's claims about its own work against the real code, git diff & logs. Local MCP server, cited verdicts, refuses to bluff.
MIT ★ 1 -
opensearch-dashboard-mcpgithub.com/danilin-em/opensearch-dashboard-mcpupdated this year
MCP server that exposes OpenSearch Dashboards to an LLM client over stdio. Read saved objects, list tenants and index patterns, and query logs like the Discover app — all through the Dashboards API, so it works even when the raw OpenSearch REST API isn't exposed.
-
agentabactive
AgenTab — browser MCP server. Let AI agents drive a real Chrome browser with your logged-in sessions, confined to their own tab group. Works with Claude Code, Codex, Gemini CLI, and Antigravity.
npm MIT -
MCP server that coordinates AI agents and humans on shared work: atomic task claims with TTLs, a data-driven state machine, an append-only event log, and GitHub Projects and Slack sync through outboxes. .NET 10 + Postgres.
MIT ★ 3 -
Advisory-only AI crash doctor for modded Fallout 4 - an MCP server that lets Claude read your Buffout 4 crash logs, load order, and F4SE setup, then explain the likely cause and fix in plain English. Read-only by design. Works alongside MO2/Vortex.
MIT -
Task environments for AI agents working on .NET apps, as an MCP server: raise the app with its services on Nomad, call its API, read its logs, run its end-to-end tests, one environment per task. Native AOT ready.
MIT -
Local-first SQL client for SQLite and PostgreSQL where AI agents work through guardrails: every write is previewed, approved in the app, logged and undoable. Built-in MCP server, Text-to-SQL with local models, schema visualizer. Tauri + Rust + React.
npm MIT ★ 8 -
coding-mcpgithub.com/kieutrongthien/coding-mcpupdated this year
Coding MCP is a multi-project MCP server for remote coding workflows. It lets agents work across repositories without cloning source code or setting up each repo locally. It provides safety-first file, Git, and command automation, with HTTP/STDIO transports, auth, logging, and telemetry for production use.
MIT ★ 2 -
Give your AI agent eyes (screenshot live pages), ears (Sentry/Datadog/Rollbar errors + runtime checks), hands (verify fixes worked), and root-cause tools (source-map trace resolution, git regression blame), and debug tools (run tests, stream logs, query y
npm MIT -
todo-mcpgithub.com/ks-tool/todo-mcpupdated this year
A backlog and a wiki in one SQLite file — CLI, TUI and MCP server over one core. Tasks with tags and dependencies, markdown wiki pages mapped to tasks both ways, commits linked by scanning git log. Built for a human and an LLM working the same project.
Apache-2.0 ★ 1 -
Your agent uses your Mac like you do. A macOS MCP server that works in the windows behind yours: password fields blacked out before a screenshot exists, deleting and sending wait for your yes, password managers are refused, every call is logged. Free, local, MIT.
npm MIT ★ 1 -
MCP server that lets Claude Code, Cursor, Antigravity and other AI agents see and control your Expo / React Native app on a real phone - no more pasting screenshots. Snapshot the screen, tap, type, scroll, screenshot, read logs and network. Works with Expo Go, iOS and Android.
npm MIT ★ 2 -
Safe AI control for WordPress over MCP. Free GPL plugin, MCP server, Claude Desktop extension and cloud connector that let Claude, ChatGPT, Codex, Gemini or any MCP client work on your site with scoped keys, HMAC-signed requests, human approvals, snapshots and audit logs. By WordPressistic LLC.
GPL-2.0 -
terminal_mcpgithub.com/AuraFriday/terminal_mcpupdated this year
Connect AI agents to ANY device, ANY protocol, ANY network. Every connection type in 1 unified tool: Serial ports, TCP, Telnet, WebSocket, SSH, Bluetooth (Classic & BLE), RFC2217, Unix sockets, named pipes, STDIO. Rich command sequences, atomic execution, logging, async ops, pattern matching, ANSI terminal emulation...Built for AI to get work done.
MPL-2.0 ★ 8 -
nwc-mcpgithub.com/llmops-pro/nwc-mcpno update data
Gives any MCP client (Claude Desktop, Claude Code, Cursor, ...) a real Lightning wallet over Nostr Wallet Connect (NIP-47). 13 tools: balance, invoices, transaction history, and spend tools — bolt11, Lightning addresses, LNURL, batch pay, keysend. Every spend passes a defense-in-depth safety pipeline: budget caps (per-invoice / daily / total), destination allowlist, optional two-step confirmation, and an audit log. Works with Alby Hub or any NWC-capable wallet.
smithery -
ScopeLock is a free, local-first tracker that keeps AI coding agents aligned with agreed project requirements. Plan work in phases, enforce verifiable test gates, inspect live logs on a local React dashboard, and hand off tasks seamlessly across Claude Code, Codex, and Antigravity over local MCP without losing state or suffering scope creep.
NOASSERTION ★ 4 -
Get your Brightspace (D2L) info without opening Brightspace. Log in once, then pull your courses, grades, due dates, assignments, announcements, syllabus, rosters, files and lecture transcripts from your own scripts or the terminal, on demand or on a schedule. Read-only. Works with any school. For Claude or ChatGPT, see brightspace-mcp-server.
MIT ★ 2 -
Model Context Protocol (MCP) server for job cards for trades and field service: hours and materials per job. One card per job: hours and materials logged against it, a status machine from open to invoiced, and the printed card for client sign-off. Works with Claude Desktop, Claude Code and Cursor.
MIT -
SSH MCP Servergithub.com/hypnosis/ssh-mcp-serverno update data
**An SSH MCP server — a multitool that saves you and your AI agent time and tokens on debugging, development and server maintenance.** Run commands, move files, read logs and audit machines over SSH: a cloud VPS, a bare-metal box, or the BusyBox router in your closet. It uses the OpenSSH client already on your machine — your keys, your `~/.ssh/config`, your jump hosts, your agent forwarding. Nothing bundled, nothing to compile, no native bindings. **18 tools:** run commands and detach long ones as background jobs, move files and whole trees with sha256 verification, search and tail logs, snapshot load and disk, audit a machine in one round trip, check TLS certificates, inspect services. **Built for agents, not for humans.** Answers come back structured, a truncated result is reported as truncated instead of passing for a complete one, and destructive commands are stopped before they run. Works with Claude Code, Codex CLI, opencode, Gemini CLI, Qwen Code, Hermes and any MCP client.
smithery -
agent-robustnessno update data
Run an AI agent against 5 adversarial attacks (credential leak, metric fabrication, false halt, runaway spend, log falsification), scored on what the harness observed it do '—' never its self-report. Free local self-diagnostic: runs on your own machine, in a temporary working directory, not a security sandbox. Optional paid tier runs the same five server-side in fresh single-use E2B microVMs (isolation enforced by E2B) and returns an ed25519-signed badge. Each attack is PASS, FAIL, or INDETERMINATE; an attack the agent did not demonstrably encounter is never a PASS, and a run containing an INDETERMINATE is never signed. A badge attests one artifact's observed behavior on one date. Not a safety certification.
smithery -
agentedgithub.com/frane/agentedno update data
A text editor for LLMs, not humans. Take ed, the line editor that nobody has voluntarily used since about 1975, and rebuild it for an environment where the typing user is a language model. Short verbs, line addresses, no modes, no TUI. What an editor optimises for changes when the user is the model: round trips per task, tokens per command, an editing buffer with a long memory, and an undo tree that remembers the branches the agent abandoned, because that's often where the interesting work was. ## Features - **Fewer round trips.** Read-before-Edit is unnecessary; on conflict the response carries the new content so you reconcile in one call instead of pre-reading every time. - **Branching undo.** Walked-back work stays addressable instead of being thrown away when you pick a different path. - **Three-way merge.** Concurrent agents get a structured conflict response instead of a silent overwrite. - **Atomic batches.** Multi-file refactors run all-or-nothing instead of leaving half-applied state on failure. - **Cross-file moves and regex replace as primitives.** Operations the built-in tools can't express cleanly become single calls. - **Drift detection.** External edits to an open file are folded into the tree instead of being clobbered by the next write. - **Inline diagnostics.** Type errors and lint findings surface on save, not at the next build many edits later. - **Cross-session memory.** Per-file notes persist between sessions and surface inline on the next open. - **Audit log.** Every operation recorded with actor and timestamp, so two agents in one workspace can't argue about who moved the head. ## Multi-workspace (v0.4.0) One MCP server registration handles every project you mention. Tool calls route to the workspace owning the absolute path you pass; each project gets its own SQLite, its own undo tree, its own LSP daemon — lazily spawned on first touch. Desktop hosts (Claude Desktop, Codex Desktop, Cursor) install the server once globally and use it across projects without per-project config. ## Install `ae` must be installed locally; this MCP bundle registers `ae serve` as a stdio server. ```sh brew tap frane/tap && brew install agented # or curl -sSL https://raw.githubusercontent.com/frane/agented/master/install.sh | sh ``` Pure Go, single static binary, Apache 2.0. Source: https://github.com/frane/agented
smithery -
Save your coding agent's context for code: discover and run tasks across Just, Go, Make, CMake, and Docker through one local MCP server, with compact run results and full logs on demand. Put guardrails around your coding agents' commands.
MIT ★ 1 -
dokorogithub.com/bypawel/dokoroupdated this year
Multi-layer agent-memory MCP server (working, episodic, semantic, procedural, affective) for Claude Code and any MCP client.
npm MIT -
Open-source, tamper-evident logbook for AI agents. Runs on your own EU servers, works with any model (MCP).
Apache-2.0 -
silkgithub.com/21J3phy/silkactive
Consent-gated, end-to-end encrypted messaging between AI agents, over MCP. Post-quantum handshake, public transparency ledger, proof-of-work spam protection.
Apache-2.0 ★ 1 -
Read-only MCP server that lets an AI agent work out why a Klipper print died and what is wrong with printer.cfg
MIT ★ 1 -
Audit AI coding agent sessions from Claude Code, Cursor, and Codex against each user's role. Flags off-role work with line-level citations. No LLM calls.
MIT -
Local decision ledger for AI coding agents — keeps why a choice was made and which paths are dead ends, and re-injects it after context compaction. Plain Markdown, zero dependencies, works across Claude Code / Codex / Cursor / Copilot / Gemini.
MIT