BingHanOfUESTC/open_agent_team

Build persistent multi-agent teams that collaborate like real organizations to deliver complex tasks.

106Stars on the repository
200Mods indexed here, across every type
3mo agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

BingHanOfUESTC/open_agent_team

Agent Claude Code

A delivery agent that organizes a research project’s code, data, environment, experiments, reports, figures, and reproduction instructions into a final checklist.

not rated 106 3mo ago A 33 tokens original MIT

environment_agent

02

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent for building a specified or reproducible research environment and recording its hardware, software versions, commands, dependencies, and smoke-test results.

not rated 106 3mo ago A 34 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

An experiment-running agent that executes research tests such as baseline comparisons, ablations, robustness checks, and sanity checks.

not rated 106 3mo ago A 30 tokens original MIT

gap_idea_agent

04

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent for turning research evidence into possible improvements and experiments. It identifies open problems, proposes ideas, and describes how to test them and how they might fail.

not rated 106 3mo ago A 37 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

A coding agent that turns a research plan into implementation code for algorithms, models, training, evaluation, data processing, and plots.

not rated 106 3mo ago A 30 tokens original MIT

latex_report_agent

06

BingHanOfUESTC/open_agent_team

Agent Claude Code

A research-writing agent that turns papers, ideas, methods, and experiment results into a compilable arXiv-style LaTeX report, with charts made from recorded data.

not rated 106 3mo ago A 38 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

A literature-research agent that searches for recent papers, existing methods, code repositories, datasets, benchmarks, and public rankings, then records where each item came from.

not rated 106 3mo ago A 38 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

A paper-reading agent that extracts evidence from important research papers, including their methods, assumptions, experiments, results, limitations, and available code or data.

not rated 106 3mo ago A 35 tokens original MIT

repo_data_agent

09

BingHanOfUESTC/open_agent_team

Agent Claude Code

A research-support agent that gathers code repositories, datasets, and pretrained model files while recording their sources, versions, licenses, checksums, and usage limits.

not rated 106 3mo ago A 31 tokens original MIT

research_plan_agent

10

BingHanOfUESTC/open_agent_team

Agent Claude Code

A research-planning assistant that turns a chosen research idea into staged experiments, resource estimates, milestones, and clear conditions for success or stopping.

not rated 106 3mo ago A 30 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

A research-scoping assistant that turns a vague research goal into a defined problem, boundaries, constraints, measures of success, and acceptance checks.

not rated 106 3mo ago A 35 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent for analyzing research or experiment results, including errors, comparisons, resource use, and how reliable the conclusions are. It also records what the results do and do not support.

not rated 106 3mo ago A 32 tokens original MIT

team_lead_agent

13

BingHanOfUESTC/open_agent_team

Agent Claude Code

A team-coordination agent for running a complete research project from defining the question through literature review, coding, experiments, iteration, paper writing, and delivery.

not rated 106 3mo ago A 53 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent that gathers the facts needed to investigate a failure, including error messages, software versions, environment details, reproduction steps, inputs, expected results, actual results, and timing.

not rated 106 3mo ago A 30 tokens original MIT

reproduction_agent

15

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent for planning a minimal reproduction: the smallest set of steps, environment details, input data, and observations needed to show a problem. A minimal reproduction helps others see and investigate a bug consistently.

not rated 106 3mo ago A 20 tokens original MIT

root_cause_agent

16

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent that develops ranked hypotheses about what caused a problem, based on available evidence. It also lists what should be checked and possible fixes.

not rated 106 3mo ago A 25 tokens original MIT

stacktrace_agent

17

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent that examines log messages, stack traces, error codes, and the code paths connected to them. A stack trace is a record showing where a program failed.

not rated 106 3mo ago A 17 tokens original MIT

verification_agent

18

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent that designs verification steps, tests, and regression checks for each identified root cause and proposed fix. The excerpt does not describe its required verification method in detail.

not rated 106 3mo ago A 23 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent for designing a business model and go-to-market plan. The description is in Chinese and lists customer problems, value propositions, competition, differentiation, and milestones.

not rated 106 3mo ago A 27 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent that reviews customer problems, evidence, business models, financial assumptions, risks, and whether a plan is clear to investors.

not rated 106 3mo ago A 29 tokens original MIT

BingHanOfUESTC/open_agent_team

Agent Claude Code

An agent that breaks product documents, customer interviews, operating data, reports, and financial tables into evidence and assumption cards. It distinguishes facts, assumptions, goals, and recommendations.

not rated 106 3mo ago A 29 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: