61,918 results in MCPs, Observability, Hooks, Evals · page 30 of 2,580
MCPsExperimental
h1-brain
Connects AI assistants to HackerOne to pull bug bounty history, program scopes, and report details into a local SQLite database, exposing tools for searching, analyzing, and generating attack briefings using both personal and public disclosed reports.
Aider MCP Server offloads AI coding tasks to Aider's specialized implementation, reducing costs and enhancing development efficiency through natural language code changes and model selection.
Phospho is a text analytics and evaluation platform for LLM apps. It logs sessions, runs clustering, detects failures, and surfaces actionable insights.
MCP server for managing video storyboard tasks, enabling Codex to automatically generate and populate images/videos based on shot descriptions and visual design specs.
Integrates with Proxmox virtualization environments to enable monitoring and management of virtual machines, containers, storage pools, and cluster resources through token-based authentication and structured API commands.
Integrates with JianYing desktop video editor using UI automation to enable draft management, media handling, track operations, and export capabilities for automated video editing workflows and batch processing.
Integrates with Home Assistant to enable natural language control of smart home devices, querying device states, troubleshooting automations, and summarizing home configurations through a Docker container that communicates with the Home Assistant API.
Integrates with Microsoft Teams to enable reading messages, creating threads, replying to messages, and mentioning team members through Azure bot credentials.
Executes local Anthropic-style skills through directory-based discovery where each skill contains YAML metadata and Markdown instructions, enabling reusable AI workflows and automation scripts with sandboxed execution environments.
Bridges Claude Desktop with macOS Messages app, enabling reading and sending iMessages through direct database access with contact lookup and group chat support.
Governance runtime for AI agents: a guard tool evaluates risky actions against policy before they execute (block / warn / require human approval), approvals route to a human queue, and every action becomes a replayable decision record with per-action spend tracking. Runs over stdio via npx @dashclaw
Integrates with Google's Gemini 2.5 Flash to generate and edit images from text prompts, supporting iterative workflows with reference images and automatic cross-platform file management.