# agentenv/agentflow

Orchestrate thousands of agents and harnesses as a graph programatically

Repository: https://github.com/agentenv/agentflow
Canonical: https://ross.abutalabs.com/products/agentenv-agentflow
Language: Python
License: MIT
License Family: permissive
Last push: 2026-08-25T20:48:03+00:00

## Health v2 (maintenance only)
Score: 59/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 12
- inputs: {"age_days": 179, "days_push": 8, "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 1377, forks 287 (observed 2026-08-28T04:04:33.368511+00:00)

## What it is
AgentFlow is a Python library for orchestrating large numbers of LLM agents (Codex, Claude, Kimi, Pi) as programmatic dependency graphs with parallel fanout, merging, and iterative cycles. Pipelines execute on local Docker, SSH, EC2, or ECS targets and can be defined in code or driven via a CLI.

## Use cases
- orchestrate thousands of llm agents in a dependency graph
- fan out parallel code review agents and merge their findings
- run iterative write-review-fix loops with llm agents until done
- build a plan-implement-review pipeline across codex and claude
- execute agent pipelines on remote ec2 or ecs targets
- run 64 parallel workers with batch merges and synthesis

## When to avoid
- you only need a single-agent chatbot or simple prompt chaining
- you want a no-code visual agent builder
- your orchestration needs are non-LLM workflow automation

## Facets
- artifact type: library
- maturity: active
- function: agent-framework, workflow-automation, llm-inference, concurrency, scheduling
- domain: large-language-models, developer-tools
- platform: python, cli, cross-platform
- tags: agent-orchestration, dag-pipeline, llm-agents, fanout, multi-agent, codex, claude, ai-agents, automation, docker

## Member repositories
- agentenv/agentflow (main) score 59

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:33.368511+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 2026-08-30T04:40:25.224809+00:00, confidence not recorded.
  - readme: https://github.com/agentenv/agentflow (fetched 2026-08-28T04:04:33.368511+00:00, sha 847eb06abe37)
- Data as of 2026-08-30T08:39:29.467469+00:00.