# polyuiislab/infiAgent

Build your own Cowork, AI Scientist and other SoTA Agents just by editing config files. Support anthropic skills. An infinite-horizon agent framework designed for long-running, complex tasks.

Repository: https://github.com/polyuiislab/infiAgent
Canonical: https://ross.abutalabs.com/products/infiagent
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: agent, framework, sdk, sdk-python
Last push: 2026-07-11T17:43:37+00:00

## Health v2 (maintenance only)
Score: 72/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 92, release rhythm 77, longevity 18
- inputs: {"age_days": 252, "days_push": 53, "days_rel": 156, "gap_med": 6, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1184, forks 52 (observed 2026-08-28T04:03:54.593701+00:00)

## What it is
infiAgent (MLA) is a Python agent framework for building domain-specific AI agents that run for days on complex tasks without context degradation. Agents are defined by editing configuration files, supporting multi-level hierarchical orchestration or flat architectures, with file-based persistent memory, crash-safe resume, and compatibility with the Agent Skills open standard.

## Use cases
- build ai agents by editing config files
- run long-running autonomous agent tasks for days
- create an ai scientist agent for research and paper generation
- recover interrupted agent runs from checkpoints
- add reusable skills to agents via the agent skills standard
- give agents persistent memory across sessions without a database
- orchestrate multi-level agent hierarchies for complex domain tasks

## When to choose
- you need agents that run continuously for days with crash recovery and no context compression degradation
- you want to build specialized agents like research assistants or coworkers purely through configuration files
- you need file-based persistent memory across sessions without external databases
- you want compatibility with the agent skills open standard and anthropic-style skills

## When to avoid
- you need a simple single-prompt chatbot with minimal setup
- you require a framework with a large plugin marketplace or desktop app bundled in this repository
- you need permissive licensing since the project is GPL-3.0
- you want a mature framework with a long production track record

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, workflow-automation, sdk, developer-tools
- domain: large-language-models, developer-tools
- platform: python, cli, cross-platform
- tags: multi-level-agent, long-running-tasks, config-driven-agents, agent-skills, persistent-memory, breakpoint-resume, multi-provider-llm, scientific-research-agent, cowork, ai-agents, automation

## Member repositories
- polyuiislab/infiAgent (main) score 72

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:54.593701+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 2026-08-30T06:24:29.332944+00:00, confidence not recorded.
  - readme: https://github.com/polyuiislab/infiAgent (fetched 2026-08-28T04:03:54.593701+00:00, sha 886441ae4e30)
- Data as of 2026-08-30T08:39:29.467469+00:00.