# rcortx/kiwiq

Production-grade multi-agent orchestration platform - JSON-defined agents, multi-tier memory, and built-in observability. Battle-tested on 200+ enterprise AI agents. Now fully open-sourced (prod at https://kiwiq.ai).

Repository: https://github.com/rcortx/kiwiq
Canonical: https://ross.abutalabs.com/products/kiwiq
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-04-13T18:33:03+00:00

## Health v2 (maintenance only)
Score: 49/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 77, release rhythm 35, longevity 12
- inputs: {"age_days": 168, "days_push": 142, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1276, forks 241 (observed 2026-08-28T04:04:10.994810+00:00)

## What it is
KiwiQ is a production-grade multi-agent orchestration platform in Python that lets you define AI workflows as JSON/Python graph schemas compiled to LangGraph and executed via Prefect. It includes multi-provider LLM support, multi-tier memory (PostgreSQL, MongoDB, Weaviate, Redis), human-in-the-loop interactions, RAG pipelines, web scraping, and built-in observability.

## Use cases
- orchestrate multiple AI agents in production workflows
- define multi-step LLM workflows as JSON graph schemas
- build RAG pipelines with vector search
- add human approval steps to AI agent workflows
- run AI-powered web scraping and lead research
- self-host a multi-agent AI platform with observability

## When to choose
- you need production-tested multi-agent orchestration with memory and observability built in
- you want code/JSON-defined workflows rather than a visual builder
- you need multi-provider LLM support and human-in-the-loop approvals
- you want ready-made workflow templates for research, content, and lead scoring

## When to avoid
- you need a lightweight single-agent library or simple LLM API wrapper
- you want a visual no-code workflow builder
- you cannot operate the required infrastructure stack (PostgreSQL, MongoDB, Redis, RabbitMQ, Weaviate, Prefect)
- you work outside Python 3.12

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, workflow-automation, rag, web-scraping, llm-inference, chatbot, monitoring, api-framework, websocket, message-queue, caching, auth, authorization
- domain: artificial-intelligence, large-language-models, backend, developer-tools, self-hosted
- platform: python, self-hosted, cloud
- tags: multi-agent-orchestration, langgraph, prefect, human-in-the-loop, multi-provider-llm, workflow-engine, json-defined-agents, fastapi, rabbitmq, weaviate, ai-agents, retrieval-augmented-generation, automation, docker, web-server

## Member repositories
- rcortx/kiwiq (main) score 49

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:10.994810+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T05:02:58.471998+00:00, confidence not recorded.
  - readme: https://github.com/rcortx/kiwiq (fetched 2026-08-28T04:04:10.994810+00:00, sha 1a5102159afd)
- Data as of 2026-08-30T08:39:29.467469+00:00.