61,920 results in MCPs, Memory, Evals, Hooks · page 29 of 2,580
MCPsPreview
f2c-figma-to-code
Bridges Figma design files to code generation, enabling direct conversion of designs into HTML, CSS, and other assets with customizable output paths and file organization.
Long-term and multimodal memory for AI agents. Store facts and conversations with add_memory, recall them with search_memories — 8 tools over stdio/SSE/HTTP. Per-character memory isolation, LLM-based semantic deduplication, FAISS + JSON storage, and a fully local option (Ollama, no API key). Drop-in
An LLM-managed knowledge base following the Open Knowledge Format (OKF) v0.1 spec. Provides MCP tools: kb_query, kb_add, kb_update, kb_status over stdio or streamable HTTP.
Enterprise multi-agent customer service system using MCP for standardized tool interfaces across order queries, ticket management, and knowledge base search.
Bridges AI with Telegram to enable monitoring, organizing, and responding to messages through account access, dialog management, and conversation handling capabilities
Provides seamless access to Apple Developer Documentation with smart search capabilities and wildcard pattern support for iOS, macOS, SwiftUI, and UIKit development, featuring priority-based framework searching and intelligent caching for efficient API discovery and symbol documentation retrieval.
Integrates DeepSeek's language models, enabling AI-powered chat completions with customizable parameters for tasks like writing assistance and code generation.
MCP control plane for Claude Code, Codex, Cursor, and OpenCode. Supports opt-in quota-aware rotation across multiple native Claude Code or Codex subscription accounts of the same harness, with isolated credentials, separate quotas, and bounded task handoffs.
Provides three sequential prompts that guide software development from high-level requirements through EARS format documentation and design specifications to final code generation, supporting a structured waterfall development workflow.
Making enterprise AI infrastructure universally accessible. Edge-first platform unifying 12 providers and 100+ models with multi-agent orchestration, HITL workflows, guardrails middleware, and context summarization.
Privacy-first SAST tool for Python, TypeScript, and Go that detects dead code, security vulnerabilities, and code quality issues with hybrid AST and optional LLM analysis.
Unifies search and content processing by dynamically selecting optimal providers like Tavily, Brave, and Perplexity to enable flexible information retrieval and enhancement across multiple domains.