# RUC-NLPIR/DeepAgent

[WWW‘26 Oral🔥] DeepAgent: A General Reasoning Agent with Scalable Toolsets

Repository: https://github.com/RUC-NLPIR/DeepAgent
Canonical: https://ross.abutalabs.com/products/deepagent
Language: Python
License: MIT
License Family: permissive
Last push: 2026-04-13T07:24:32+00:00

## Health v2 (maintenance only)
Score: 52/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 77, release rhythm 35, longevity 24
- inputs: {"age_days": 338, "days_push": 142, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1134, forks 140 (observed 2026-08-28T04:03:43.074338+00:00)

## What it is
DeepAgent is a general reasoning agent framework that dynamically searches for and uses tools from large toolsets (e.g., 16,000+ RapidAPIs) in an end-to-end agentic reasoning process. It is a Python research codebase from RUC NLPIR accompanying a WWW 2026 paper, deployable with reasoning models like QwQ and Qwen3.

## Use cases
- build an LLM agent that discovers and calls APIs from a large toolset
- run a reasoning agent on general tasks with scalable toolsets
- evaluate agents on ToolBench-style API tool use
- run embodied AI agent tasks in ALFWorld
- deploy an agent with QwQ or Qwen3 reasoning models

## When to choose
- you need an agent that scales to thousands of tools via retrieval
- you are reproducing or extending the DeepAgent research
- you want an open-source reasoning agent with custom toolsets

## When to avoid
- you need a production-ready commercial agent platform
- you want a simple agent without tool retrieval complexity
- you need non-Python environments

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference
- domain: artificial-intelligence, large-language-models
- platform: python, windows
- tags: tool-retrieval, rapidapi, agentic-reasoning, research, llm-agent, tool-use, reasoning, ai-agents, linux, macos

## Member repositories
- RUC-NLPIR/DeepAgent (main) score 52

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:43.074338+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 2026-08-30T06:36:53.066152+00:00, confidence not recorded.
  - readme: https://github.com/RUC-NLPIR/DeepAgent (fetched 2026-08-28T04:03:43.074338+00:00, sha 085a56f06203)
- Data as of 2026-08-30T08:39:29.467469+00:00.