MCP servers that give coding agents real developer tooling: linters, package registries, test runners, language docs and local build commands. 13,969 servers in this category.
Allows an AI client to query business metrics in plain English using a semantic layer to ensure accurate, SQL-free data retrieval.
Provides an exact arithmetic, symbolic math, statistics, and SQL-over-files engine for local language models with auditable computation logs.
Provides a persistent, workspace-scoped Python runtime that allows LLM agents to execute code, manage files, and run shell commands.
Allows AI clients to monitor real-time model quotas and switch between Google accounts within the Antigravity IDE.
Enables heterogeneous coding agents to perform adversarial peer reviews through a shared, append-only audited governance board.
Provides coding agents with release histories and change logs for packages across PyPI, npm, crates.io, NuGet, and other major registries.
Evaluates data visualizations and provides statistical evidence to guide machine learning actions and feature engineering.
Compresses JSON payloads into Token-Optimized Object Notation (TOON) to reduce LLM context window consumption by up to 60%.
Enables AI assistants to manage Python environments, install packages, and audit dependencies using the uv package manager.
Provides persistent memory and issue tracking for AI coding agents to manage todos, changelogs, plans, and session handoffs across projects.
Lets an AI client capture screenshots, record screen video, and extract frames from recordings on Windows.
Connects AI clients like Claude Desktop and Codex to a running GNU Emacs instance via emacsclient to manipulate buffers and files.
Enables AI coding agents to perform surgical, byte-correct code edits using Abstract Syntax Trees instead of brittle search-and-replace operations.
Provides a multi-agent council for coding agents to get independent, read-only second opinions on code reviews and plans.
Connects Autodesk Revit to AnythingLLM to enable RAG-based architectural auditing and deterministic QA/QC against local document libraries.
Allows AI assistants to lint, format, and automatically fix Python code using the Ruff toolset.
Provides AI assistants with tools to lint, format, and statically analyze PHP files and directories using the Mago engine.
Enables an AI client to store and retrieve verified bug fixes and debugging preferences using a local SQLite database.
Wraps REST APIs as MCP tools with Pydantic validation and error handling for use in AI clients like Claude and Cursor.
Provides deterministic verification for AI assistants by executing Python code in a restricted environment to verify math, logic, and code.
