Development MCP Servers

Public remote MCP for generator guide search, connector compatibility, and extension-cord sizing.

Set up GDPR and CCPA cookie consent: create a site, publish a banner, scan and categorize trackers.

One searchable memory across Claude Code, Gemini, Codex, OpenCode, Antigravity and Cursor sessions.

Design the world, tune the balance, and plan the work — together. IMS Creators brings your GDD, game data, kanban boards, and people into one live cloud workspace, in any browser.

航班管家 DAST 官方航空数据 MCP，提供航班动态、延误预测、机场天气、飞行轨迹及航班舒适度等数据能力。

Symvanta gives your AI coding agent your codebase's real call graph over MCP, so it stops guessing and knows what breaks before it edits.

AI-native spec-first workflow tool: project registry, behavioral specs, and engineering threads over MCP

Model Context Protocol server for Matrix42 — lets AI assistants explore the API, read the data model, search the service desk, and act on tickets. Read-only by default.

JSON formatter, validator, and repair tool. Accepts JSON5, JavaScript objects, and Python dicts (single quotes, None/True/False, tuples).

Context optimization for LLM agents: tool registry, result masking, budgeting, compaction.

Stream live EEG brain state (focus, calm, attention) from any EEG device into Claude and any MCP client — a real Model Context Protocol server + brain-computer interface toolkit.

Gives AI assistants a shared, live view of your terminal session for debugging CLIs and TUIs, or autonomous terminal control.

MCP (Model Context Protocol) server that exposes the fetch-cwe-list library as tools for LLM agents like Claude.

Independent review layer for AI-generated text: a separate model runs a fixed ten-check gate over output it did not write and returns the flagged line and the reason per check, never a grade.

An AI-native WordPress theme with a built-in MCP server that gives an agent 34 scoped abilities over pages, templates, design tokens, menus, SEO and WooCommerce, secured with OAuth and a one-click kill switch.

Grounded AI code generation for Microsoft Dynamics 365 Finance & Operations (X++).

Nine read-only MCP tools for evidence-backed open-source AI project research, license checks, comparisons, alternatives, and stack planning.

Local-first self-hosted runtime and stdio MCP bridge for auditable AI agent sessions.

Pay-per-use tools for AI agents, with one prepaid balance and no separate provider API keys.

MCP server exposing NIST NVD CVE search, CPE search, and CVSS v2/v3/v4 calculators as tools.

Evidence, authorization, recovery, and signed traces for MCP-capable AI agents.

MCP server that gives AI coding agents a deterministic test loop for the MCP servers they build: write acceptance tests in natural language, validate the wording, run them, and read structured pass/fail results — no AI interprets the tests at runtime. Run with: npx mcprigor serve <project>. Also a full CLI/CI test framework with contract drift detection and transport parity. Apache-2.0.

Crash-safe task tracking and coordination for AI coding agents — expiring lease claims, resumable attempts, version-pinned reviews; one Go binary with SQLite.

Anchor a file's SHA-256 fingerprint to Bitcoin via OpenTimestamps — the file never leaves the device.

Scans source code for PHI leaking into LLM prompts, logs, and analytics calls - local-first, no network calls.

▎ A code graph that extends to any language via pluggable adapters, with task/QA traceability and real MCP tools — zero-config to try. ▎

MeMesh 是跑在你自己電腦上的 AI Agent 協作層。記憶、指定收件人的訊息和改善提案，都放在同一個 SQLite 檔案裡。Claude Code、Codex、Cursor、Gemini，或你自己接的本機模型，都能透過 MCP、HTTP 或 CLI 使用；不綁單一廠商，也不用先把資料送上雲端。開源、MIT 授權、免費使用。

The Otto MCP Server connects your LLM directly to Otto Webmaster, enabling external AI clients to execute CMS content updates, run automated SEO hygiene, monitor site health, and fetch analytics directly within your agent workflows.

Drives the browser you are already signed into, so work behind a login, SSO or MFA needs no scripted auth — 20 tools standalone, 99 with the free T3rnel Browser extension.

READ-only today by design; WRITE is staged behind explicit safety architecture