MCP servers that give coding agents real developer tooling: linters, package registries, test runners, language docs and local build commands. 13,961 servers in this category.
Audits regular expressions for catastrophic backtracking vulnerabilities using character-set algebra to identify potential DoS vectors.
Provides reusable tooling, project templates, and scripts to manage Claude Code workflows, project records, and skill deployment.
Provides local observability, usage analytics, and error debugging for AI coding agents and local LLM services via a SQLite-backed dashboard.
Scans AI-agent skills and MCP servers for security risks using deterministic rules without executing code or using an LLM.
Acts as a runtime firewall for AI agents to prevent tool poisoning, data theft, AI worms, and memory poisoning in MCP traffic.
Detonate and scan MCP servers in a sandboxed environment using canary credentials to detect malicious behavior before installation.
Identifies breaking changes when migrating Python MCP servers from SDK v1 to v2 by scanning code with AST analysis.
Reduces Claude Code token usage by running tests, git operations, code reviews, and project searches using local models via Ollama.
Synchronizes instructions, MCP servers, and skills across multiple AI coding agents like Claude Code, Codex, and OpenCode.
Provides hands-on Python projects and theory to teach AI engineering, covering agent loops, RAG, LangGraph, and MCP through DevOps-focused use cases.
Build Model Context Protocol servers in pure Python using a zero-dependency framework.
Performs static security analysis on AI agent codebases to identify MCP integrations, credential exposure, and potential attack paths.
Compiles natural language into typed execution graphs and DAGs to prevent prompt injection and hallucinated tool calls in autonomous agents.
Audits code against legal standards, generates fixes, and uses Phoenix traces to improve compliance scans.
Performs heuristic security audits on MCP servers and client configurations to detect tool poisoning, over-privileged capabilities, and leaked secrets.
Tests whether an MCP client correctly handles OAuth token expiration by using refresh tokens with the required resource parameter.
Orchestrate a team of AI agents using desktop-based queues, pipelines, and live terminals with human-in-the-loop capabilities.
Explains smart contract functions using official documentation and executes read-only on-chain calls on the Pharos Pacific Mainnet.
Unifies conversation history across multiple AI coding agents by grouping sessions into project folders for cross-agent context sharing.
Analyzes the token cost and selection accuracy of MCP server stacks to prevent model confusion and optimize tool usage.
