Search

AI Power Grid text, image, video, and audio generation for n8n

open-source tools for agentic research

Gate OpenAI function/tool calls through Themis

Gate LangChain tools (and anything tool-shaped) through Themis

vLLM (OpenAI-compatible) LLM client for AI-Parrot (parrot.clients.vllm)

OpenRouter LLM client for AI-Parrot (parrot.clients.openrouter)

Provider-neutral compute transfer client for Central Storage Platform

Probe any ACP-compatible agent for its capabilities (models, modes, configOptions, prompt capabilities, auth methods, MCP transports). No real prompt required; no token cost.

Score a RAG system against a golden dataset and fail the build when it gets worse. Deterministic metrics, no LLM judge, no API key, no per-run cost. Zero dependencies.

Z.ai (Zhipu) LLM client for AI-Parrot (parrot.clients.zai)

Athena LLM client — one seam across frontier, hosted-open (with failover), and local backends

OpenAI-compatible embeddings and chat for Backstage RAG. Works with OpenAI, Ollama, Gemini and any compatible endpoint.

The s3:// storage backend for dirigent, compatible with any S3 API.

Real-time web dashboard for monitoring and controlling multiple Claude Code sessions

Advanced GraphQL n8n node for Twenty CRM with custom database, field support, and resource locators. Complete CRUD support for custom and default databases

Not publicly released yet. Install the beta: npm i @cognition-ai/sdk@beta

Wire-format adapter that exposes Vercel AI SDK provider instances over the OpenAI HTTP v1 API as a standard fetch handler. Transport-agnostic — no Hono / Express / Node http dependency.

n8n node for Basecamp 4 — projects, to-do lists, to-dos, messages, campfire, comments, people, attachments

Out-of-tree speculative decoding plugin for vLLM-HUST on Ascend

Run OpenAI Guardrails inside the Vercel AI SDK, as language model middleware. Works with any AI SDK provider.