Search

ローカルLLM（mlx / mlx-vlm / llama.cpp / router）を OpenAI 互換 API として起動・管理する軽量サーバー

No description available

Core RAG primitives: Chat orchestrator, provider interfaces, and Settings registry

A fail-closed verification layer you wrap around your LLM: it never lets an unsupported claim through as ACCEPT, re-derives what is re-derivable, and emits a deterministic verifiable reward.

Git for prompts - version, diff, and test your LLM prompts across any model

Wallet connector implementation using entity storage

Budget: pre-flight cost caps, spend attribution, and circuit-breakers for LLM calls.

MemGPT memory backend for OpenClaw — three-tier memory architecture via pymemgpt sidecar

[](https://www.npmjs.com/package/@agentclientprotocol/codex-acp)

Claude Code skill: port HTML/CSS/JS layouts onto a custom WordPress theme

Vercel AI SDK provider for OpenGradient TEE LLM (verifiable inference)

Minimal skill-based agentic SDLC framework

Befroosh custom n8n nodes (SMS providers, etc.)

Terminal benchmark runner with automatic discovery of local LLM servers

A CLI and SDK tool to audit RAG pipelines for systemic infrastructure weak points.

n8n community node for Morning (Green Invoice) — Israeli accounting and invoicing API

Helper module for the FAIRSCAPE RO-Crate wizard Claude Code skill bundle

n8n community node for Viber Bot API (https://github.com/andrey0001/n8n-nodes-viber-bot)

n8n node for signing Solana transactions via @solana/keychain backends (Memory, Vault, AWS KMS, GCP KMS, Privy, Turnkey, Fireblocks, and more)

PREVIEW (not part of the supported v1 content seam) — build-time embedding generation for semantic search. A fresh build environment downloads the all-MiniLM-L6-v2 model (~25MB via @huggingface/transformers) on first run; cache the model dir in CI. Experi