# Emiyaaaaa/HiveMind

A Python-first runtime layer for multi-agent systems, with persistent run state, streaming execution events, and a pluggable orchestration interface.

Repository: https://github.com/Emiyaaaaa/HiveMind
Canonical: https://ross.abutalabs.com/products/emiyaaaaa-hivemind
Language: Python
License: NOASSERTION
License Family: other
Last push: 2026-09-02T07:19:38+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 35, longevity 26
- inputs: {"age_days": 370, "days_push": 0, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1092, forks 129 (observed 2026-09-03T02:15:05.754238+00:00)

## What it is
AgentFlow (repo HiveMind) is a Python-first runtime layer for multi-agent systems that adds persistent run state, ordered step/message/tool-call records, and SSE event streaming on top of orchestration frameworks like LangGraph and PydanticAI. It pairs a Java/Spring Boot HTTP API with Python workers, SQLAlchemy persistence, Redis job queues, and a Next.js admin console for inspecting runs.

## Use cases
- run multi-agent workflows with durable run history across restarts
- stream agent execution events to web clients via SSE
- switch between LangGraph, AutoGen, or CrewAI without changing the API
- debug and audit agent tool calls and messages
- inspect active and historical agent runs in an admin console
- cancel, retry, and resume agent runs in production
- move an agent prototype to an inspectable service

## When to choose
- you need production runtime concerns (persistence, streaming, cancellation) around an existing agent framework
- you want framework-agnostic orchestration adapters with a stable API and schema
- you need an operator-facing console for run observability

## When to avoid
- you only need local agent composition for prototypes without persistence or streaming
- you want a single-framework solution fully managed by LangGraph, AutoGen, or CrewAI
- you cannot operate a multi-language stack (Java, Python, Redis, Next.js)

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, workflow-automation, streaming, monitoring, api-framework, message-queue, database, webhook
- domain: large-language-models, artificial-intelligence, developer-tools, backend, apis, microservices
- platform: python, self-hosted, cross-platform
- tags: multi-agent-runtime, orchestration-adapters, langgraph, pydantic-ai, run-persistence, sse-events, spring-boot, redis-queues, admin-console, agent-observability, ai-agents, docker, web-server

## Member repositories
- Emiyaaaaa/HiveMind (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:05.754238+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 2026-08-30T06:49:57.093911+00:00, confidence not recorded.
  - readme: https://github.com/Emiyaaaaa/HiveMind (fetched 2026-09-03T02:15:05.754238+00:00, sha 6a9f0e7a07af)
- Data as of 2026-08-30T08:39:29.467469+00:00.