MCP servers that give coding agents real developer tooling: linters, package registries, test runners, language docs and local build commands. 13,958 servers in this category.
Provides a preview, apply, and undo workflow for state-changing MCP tools using a durable SQLite ledger.
A coding agent that fixes failing tests within a secure, network-isolated Docker sandbox using local LLMs.
Allows AI coding agents to transfer reviewed context, instructions, and work history between different development environments and harnesses.
Provides local semantic and keyword search for coding agents across notes, PDFs, docx files, and codebases.
Audits GitHub Actions workflows for security vulnerabilities using 37 deterministic rules and provides evidence-based findings.
Statically reviews AI agent instruction files and MCP manifests to detect security risks like credential leaks, destructive commands, and prompt injection.
Enables AI agents to validate data accuracy through CSV health scoring, date verification, and schema inference.
Turns Markdown directories into an addressable node graph that AI agents can search, read, and verify via MCP.
Provides debugging tools and a dashboard for Linksys Velop mesh networks to monitor node health, sync status, and network topology.
A stdio proxy that keeps MCP servers dormant until a tool is called, significantly reducing memory usage for Claude Code and Codex sessions.
Automates the configuration of MCP servers for editors like Claude and Cursor while keeping API keys securely in an encrypted vault.
Allows developers to build MCP servers that expose a command tree through two tools, help and run, to minimize context window usage.
Performs headless smoke testing for MCP servers by validating tool schemas and executing tool calls with generated arguments.
Optimizes Antigravity 2.0 runtimes by multiplexing hooks, healing Git worktree deadlocks, and sanitizing MCP tool identifiers.
Acts as a security gateway between coding agents and MCP tools by enforcing tool registries, taint tracking, and human approval for sensitive actions.
Provides a zero-dependency proxy to fix non-compliant JSON schemas and handle OAuth authentication for the COROS official MCP in DeepSeek Harness.
Provides headless CAD agents with a deterministic 3D-printing critic to audit STL files for topology, manifold errors, and printability issues.
Tests MCP server authorization compliance by probing for forged tokens, audience binding issues, and session replay vulnerabilities.
Vet AI agent skills, MCP servers, plugins, and rules files for hidden instructions, malicious setup steps, and security risks before installation.
Scans MCP configuration files to identify stale, duplicated, or over-broad tool wiring while ensuring secrets remain redacted.
