AI & Agents MCP Servers for OpenAI Agents SDK

Run local prompt regression checks from prompt-response logs in AI coding assistants.

Finds a plan's blind spots: unstated assumptions, omitted risks, and missing pieces.

Local stdio MCP server for the AEGIS governance client. The hosted evaluation service is offline.

Rate-limited MCP HTTP proxy: sliding window plus token bucket pacing so long tasks avoid 429 errors.

Agent-to-agent communication bridge for multi-agent systems

Real-time communication between Claude Code instances across machines and workspaces. Redis-backed channels with background task

Fail-closed quality gate and hash-chained receipt ledger for AI agent workflows.

Agent-harness picks, decision guides, setup templates, and skills; live GitHub/HN discovery.

Inspect AgentMark traces and drive AgentMark Cloud: apps, experiments, metrics, deployments.

Fan a goal out to cheap parallel LLM workers on your own key or a free local Ollama. Keep your mind.

The immutable delegation frame for AI agents: gate any sensitive action, human-in-the-loop.

FIBER decision-context compiler with honest omission accounting: 259 evidence-bearing MCP tools.

AI routing, memory, guardrails, and governance. Routes across Claude, GPT, Gemini.

MCP bridge for inter-session communication between Claude Code instances

Recursive Claude spawning for autonomous project builds with self-healing test loops

Search agent skills ranked by measured lift vs a no-skill baseline, with safety and rankings.

MCP server for grounding / hallucination checking: verify a claim against evidence (Supported/Refuted/Not Enough Evidence) — for AI agents.

AI Agent Guardrails MCP server - security layer

Orchestrate multiple terminal AI CLIs (Claude Code, Codex...) into an office: employees in split real windows collaborating via an MCP bus,

MCP server for a federated AI-commons governance simulation with a verified compliance oracle.

High-performance, modular MCP v2 server with safe profiles, worker pool, caching, and stdio.