# SkyworkAI/DeepResearchAgent

DeepResearchAgent is a hierarchical multi-agent system designed not only for deep research tasks but also for general-purpose task solving. The framework leverages a top-level planning agent to coordinate multiple specialized lower-level agents, enabling automated task decomposition and efficient execution across diverse and complex domains.

Repository: https://github.com/SkyworkAI/DeepResearchAgent
Canonical: https://ross.abutalabs.com/products/deepresearchagent
Homepage: https://skyworkai.github.io/DeepResearchAgent/
Language: Python
License: MIT
License Family: permissive
Topics: gaia, general-purpose, multiagent-systems, multimodel
Last push: 2026-05-04T09:15:39+00:00

## Health v2 (maintenance only)
Score: 68/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 80, release rhythm 72, longevity 33
- inputs: {"age_days": 470, "days_push": 121, "days_rel": 190, "gap_med": 0, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3532, forks 453 (observed 2026-08-28T04:08:08.966007+00:00)

## What it is
DeepResearchAgent is a Python framework for hierarchical multi-agent systems that combines a top-level planning agent with specialized lower-level agents for deep research and general-purpose task solving. It provides a self-evolution protocol and runtime with protocol-registered resources (prompts, agents, tools, environments, memory) and a closed-loop optimization layer for iterative agent improvement.

## Use cases
- build a hierarchical multi-agent system for deep research
- automate task decomposition across specialized agents
- run a deep research agent on the GAIA benchmark
- self-improve LLM agent prompts and solutions iteratively
- orchestrate tool-calling agents with memory and tracing
- evaluate agents in browser, filesystem, or trading environments

## When to choose
- you need coordinated multi-agent planning and task decomposition
- you want an evolvable agent runtime with versioning, tracing, and rollback
- you need composable agents, tools, environments, and memory in one Python framework
- you are benchmarking general-purpose agents on tasks like GAIA

## When to avoid
- you need a simple single-agent chatbot with minimal setup
- you want a production-ready managed agent service rather than a framework
- your use case is unrelated to LLM-based agents or research automation

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, prompt-engineering, rag, workflow-automation, web-scraping
- domain: artificial-intelligence, large-language-models, developer-tools
- platform: python, cli, cross-platform
- tags: multi-agent, hierarchical-agents, self-evolution, deep-research, gaia-benchmark, tool-calling, agent-memory, llm-optimization, ai-agents, automation

## Member repositories
- SkyworkAI/DeepResearchAgent (main) score 68

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:08.966007+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 2026-08-29T18:35:05.101081+00:00, confidence not recorded.
  - readme: https://github.com/SkyworkAI/DeepResearchAgent (fetched 2026-08-28T04:08:08.966007+00:00, sha 0415866ac77a)
- Data as of 2026-08-30T08:39:29.467469+00:00.