# i-am-bee/agentstack

Deploy, and share agents with open infrastructure, free from vendor lock-in.

Repository: https://github.com/i-am-bee/agentstack
Canonical: https://ross.abutalabs.com/products/i-am-bee-agentstack
Homepage: https://agentstack.beeai.dev
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-26T16:24:33+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 77, longevity 41
- inputs: {"age_days": 583, "days_push": 7, "days_rel": 156, "gap_med": 12, "n_releases_24m": 20}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1149, forks 185 (observed 2026-08-28T04:03:46.353500+00:00)

## What it is
Agent Stack is open-source infrastructure (hosted by the Linux Foundation) for turning AI agents into running HTTP services, built on the Agent2Agent (A2A) Protocol. It bundles an agent runtime, LLM routing across 15+ providers, vector storage, authentication, S3-compatible file storage, a CLI, a web UI, and a Kubernetes Helm chart so agents can move from local development to production without rewrites.

## Use cases
- deploy ai agents as backend services
- run agents locally while developing then ship to production
- self-host an agent runtime without vendor lock-in
- expose langgraph or crewai agents over http via a2a
- add llm routing and vector search to my agent app
- manage agent secrets and oauth integrations with slack or google drive
- build a custom chat ui for my agents with a typed sdk

## When to choose
- you want to run agents as production services without building deployment infrastructure yourself
- you need framework-agnostic agent hosting (LangGraph, CrewAI, custom code)
- you want an open, A2A-compatible, self-hostable alternative to proprietary agent platforms
- you need built-in LLM provider routing, embeddings, file storage, and secrets management

## When to avoid
- you only need a lightweight agent library inside a single app with no service deployment
- you want a fully managed hosted platform rather than self-hosting a VM/Kubernetes stack
- your environment cannot run Lima/QEMU VMs or Kubernetes locally
- you need a mature, long-proven product - the project is young and Windows support is experimental

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, rag, deployment, cli, secrets-management, file-upload, mcp
- domain: large-language-models, developer-tools, self-hosted, infrastructure-as-code
- platform: windows, python, cli, self-hosted
- tags: a2a-protocol, agent-deployment, agent-runtime, llm-routing, vector-storage, helm-chart, linux-foundation, beeai, ai-agents, linux, macos, docker, kubernetes

## Member repositories
- i-am-bee/agentstack (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:46.353500+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 2026-08-30T06:33:44.565607+00:00, confidence not recorded.
  - readme: https://github.com/i-am-bee/agentstack (fetched 2026-08-28T04:03:46.353500+00:00, sha 08d1540d3a28)
  - homepage: https://agentstack.beeai.dev (fetched 2026-08-29T12:38:43.846001+00:00, sha 758ac4934d4b)
  - site_page: https://agentstack.beeai.dev/stable/introduction/quickstart (fetched 2026-08-29T12:38:43.855269+00:00, sha 64c82b8966e7)
  - site_page: https://agentstack.beeai.dev/stable/custom-ui/getting-started (fetched 2026-08-29T12:38:43.857181+00:00, sha 920c283c4bc4)
- Data as of 2026-08-30T08:39:29.467469+00:00.