agent-infrastructure
16 servers · 882★ total
Portable semantic memory for AI agents: core engine, TypeScript SDK, framework adapters, MCP server, CLI, and host plugins.
Open-source control plane and runtime for organisational agents: shared company context, isolated execution, approvals and MCP.
OpenHermit is the open-source platform for deploying fleets of AI agents as production services — durable state, sandboxed execution, managed at scale, and the channels you already use.
Give your AI agent a wallet, email, phone number, and instant access to paid MCP tools — web search, image/video/music gen, X search, SMS, voice calls, code execution, and more. Works with Gemini CLI, Claude, and any MCP client.
Verifiable and free cloud compute for AI agents. webMCP + MCP native. Check out our sandboxed Beta + research in the README
cmux MCP server — 42 tools for terminal agent lifecycle management. Spawn, monitor, and orchestrate Claude, Codex, Cursor, Gemini, Kiro agents. Thinking-state detection, chunked input, layout management.
OpenRouter for agents and sandbox environments. Provides a control plane and configuration options for persistent agent workers. Route runtime + substrates, provision specialist presets, and supervise your fleet.
Three packages: @kernel.chat/agent-os (POSIX for AI agents — capabilities, namespaces, quotas, taint, audit, vault, outcomes), @kernel.chat/kbot (terminal AI agent, MCP-native, BYOK), @kernel.chat/kbot-finance (audit-grade AI for regulated industries). Provenance-engineering substrate.
Search infrastructure for AI agents: free by default, MCP-ready, LLM-context aware, and production-grade when needed.
Governed state engine for AI agents: typed, executable knowledge artifacts with a code-like lifecycle
MCP-native infrastructure for persistent AI agent memory with encrypted storage, semantic retrieval and cross-agent continuity.
Execution control layer for AI agents — prevents duplicate or incorrect real-world actions under retries, uncertainty, and stale context.
Deterministic MCP tool flows for AI agents. Remove unnecessary LLM calls between predictable tool steps.
Model Context Protocol SDK for building structured, production-grade AI systems.
A self-hosted, containerized platform for AI agents, exposed as Capability Packs — schema-validated, one-shot JSON tools — and native MCP. The defining metric is ≥90% pack success on 7B–30B-class open-weight models, something no frontier-targeting competitor is optimizing for.
OmniBridge is an MCP server that gives AI agents an ephemeral, gVisor-isolated sandbox to run, test, and cryptographically prove their code works — before it ever reaches production.