ai-infrastructure — independent software & tools
-
agentforge-runtime
AgentForge Protocol + Cloud: create, run, monetize, evaluate, and operate AI agents across APIs, marketplaces, workflows, channels, and enterprise operations.
-
ollama-mesh
The control plane for self-hosted AI inference. Warm-state GPU routing, multi-runtime orchestration across Ollama, vLLM, llama.cpp, and TGI. Single Go binary. Apache-2.0.
-
clickllm
Prove which open model can replace your closed one — on your traffic, your hardware, your budget. Then move you there without a risky cutover.
-
distil
Compression with a quality contract — cache-aware, causally-pruned LLM context compression for agentic runtimes, certified non-inferior across 7 domains. Works with any SDK.
-
mem
A portable, self-hosted memory plane for AI agents — one core across API, MCP, CLI, and UI.
-
harnessrouter
HarnessRouter Community Edition: the self-hosted, Apache-2.0 edition of the unified interface for agent harnesses. Run Codex, Claude Code, Hermes, and more through one API, with sessions, streaming, files, cancellation, and failure handling. Implements the Unified Harness Protocol (UHP), an open standard. Your keys, your infrastructure.
-
mcp-servers-live
Auto-updated index of MCP servers shipping on GitHub, refreshed every 15 minutes
-
agentic-harness
Self-hosted completion gate for coding agents: independent verification before done, durable evidence, CLI and local GUI.
-
agentic-mesh
Multi-Agent AI Orchestrator 2026 🚀 | YAML, 6+ LLM Providers, ReAct & Swarm
-
Open-Arsenal
Open Arsenal specs can be self-hosted on private or public networks, with free use under AGPL-3.0+ requiring Section 7 attribution. Optional Specification Branding Licenses allow use without attribution, with pricing based on spec, scope, and network size.
-
memql
AI-native time-series memory graph with a single DSL — unifies concepts, queries, agent workflows, and voice into deployable primitives.
-
claude-memory
Enable persistent memory for Claude that works across crashes and restarts in both Claude Code and Claude Cowork environments.
-
unified-ai-system
Terminal-first, self-hosted AI gateway with 8 MCP tools for Codex, Cursor, and Cline. One Docker command, no API key.
-
inkbox
SDK, skills, and examples for Inkbox: give AI agents a persistent identity with email, phone number, and internet address
-
DIO
Drop-in OpenAI- and Ollama-compatible LLM gateway that learns each backend's latency online and routes vLLM / SGLang / TGI / Ollama with SLO-aware admission. No engine patches.
-
cli_subprocess_core
Foundational Elixir runtime library for deterministic CLI subprocess orchestration, normalized event and payload semantics, provider profile contracts, registry plumbing, and shared support modules for the nshkr AI SDK runtime stack rearchitecture.
-
neva
Model Context Protocol SDK for building structured, production-grade AI systems.
-
rorschach
Intelligence without consciousness. Process without self. Scale without limit.
-
systemprompt-core
AI governance infrastructure for agentic systems. Rust library behind systemprompt.io — MCP, A2A, OAuth2, audit trails, compile-time extensions. Evaluate with systemprompt-template.
-
Ops-Center-OSS
Ops-Center — the control plane for a Sovereign Zero-Trust Federated Mesh. Identity, orgs, entitlements, billing, and a local-first inference gateway for the whole Unicorn Commander suite.
-
agentic-fleet-hub
Self-hosted orchestration layer for autonomous AI agent teams. Shared memory, heartbeat scheduling, vault-first secrets, and cross-model peer review — one command to deploy.