fashionmascherine-svg/formalswarm 预览 preview

fashionmascherine-svg/formalswarm

Multi-agent validation for any repository: independent theses, adversarial critics, and a seal whose verdict is computed from real command exit codes — never from an agent's prose. One body, three runtimes: DeepSeek Harness, Claude Code, ZCode.

Project Overview项目介绍

FormalSwarm is an adversarial multi-agent validation tool that works across multiple AI coding agent platforms, including DeepSeek Harness, Claude Code, and ZCode. It structures collaboration between different agent roles: theses agents propose solutions, antithesis agents challenge assumptions, and seal agents run pre-defined verifiable commands. A deterministic rollup process calculates a final verdict of CONFIRM, REVISE, or INCONCLUSIVE based on actual check results and exit codes, rather than relying on vague AI confidence scores. It can be used for code review in existing software repositories, as well as guided document-driven problem solving for non-software use cases.

To start a workflow, users first prepare a workspace folder with carefully crafted all relevant context documents, including objectives, existing evidence, project constraints, and current background information. After installing the tool, users explicitly ask their connected agent to run FormalSwarm on the prepared folder, specifying the number of agents for each role and the detailed scope of the debate. It supports workflows ranging from small 6-agent pilot tests up to large-scale validation involving hundreds of agents, making it suitable for developers, product designers, and project planners who need rigorous validation of carefully crafted AI-generated proposals.

FormalSwarm requires Node.js version 18.17.0 or higher to run, and is released under the open-source MIT license, which means it is free to use, modify, and distribute for both personal and commercial projects. It has several key limitations to note: it does not include built-in specialist tools for specific domains, cannot automatically correct AI hallucinations, and does not replace your existing continuous integration workflow. All verifiable commands must be pre-defined by the user, and contributors are welcome to submit issues with counterexamples that return incorrect verdicts to help improve the open-source project's rollup logic.

FormalSwarm是一个多智能体对抗验证工具,可在多个AI编码代理平台上运行,包括DeepSeek Harness、Claude Code和ZCode。它的核心功能是让多个智能体分工协作,通过命题提出、对抗质疑和可运行验证三步流程,对解决方案或代码变更进行系统性审查,最终输出明确的确认、修订或无结论的裁决。它既可以用于代码仓库的代码审查,也能处理基于文档的非代码问题,比如营销漏斗设计、项目方案压力测试等。

典型使用流程是用户准备好包含目标、约束条件、现有信息等内容的工作文件夹,安装工具后,调用FormalSwarm让多个代理按既定流程展开辩论和验证。从6个智能体的小规模试点到上百个智能体的大规模验证都可支持,适合需要对AI提出的方案进行严谨校验的开发者、产品设计师和项目规划人员使用。

该工具基于Node.js开发,要求Node版本不低于18.17.0,采用MIT开源许可,可免费使用和修改。它本身不提供内置的专业领域工具,也不能保证自动修正AI的幻觉,不会替代现有CI流程,所有可运行验证都需要用户预先定义。首次使用需要按文档准备上下文文件,明确验证目标后再调用工具。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 1 warning1 项注意
  • Only 2 stars - very few users, little community feedback星标只有 2,几乎没人在用,遇到问题缺少社区反馈
DSH walks through these 9 checksDSH 会逐条核对这 9 项

Compatibility兼容性

  • DSH, Node, OS and profile requirementsDSH 版本 / Node 版本 / 操作系统 / profile 是否满足要求
  • External dependencies and runtimes (Electron / Python / Docker, ...)外部依赖与运行时(Electron / Python / Docker 等)是否齐备
  • Conflicts with installed plugins: command names, skill / tool names, ports, duplicate MCP registration与已装插件是否冲突:命令名、skill / tool 重名、端口占用、重复 MCP 注册

Security安全性

  • Repo matches the facts registered here; archived or abandoned?仓库是否与页面登记一致,是否归档或长期停更
  • Safety of preinstall / install / postinstall and install.sh / setup.ps1preinstall / install / postinstall 与 install.sh、setup.ps1 是否安全
  • curl|bash, download-then-execute, obfuscation, unrelated domains → stop immediatelycurl|bash、下载即执行、混淆代码、无关域名 → 立刻停止
  • Typosquatting or unmaintained packages among the new dependencies新增依赖里有没有 typosquatting 或无人维护的包
  • Requested permissions vs. what the feature actually needs申请了哪些权限、是否超出功能所需(filesystem / network / shell / clipboard)
  • Any sudo / admin requirement, plus uninstall and rollback是否要求 sudo / 管理员权限,以及卸载与回滚方式

Anything uncertain must be marked unknown with a note on how to confirm it. This site's signal screen is a static snapshot, not a security audit.拿不准的必须标「未知」并说明要我怎么确认。本站的信号筛查是静态快照,不能替代安全审计。

Or use CLI install (for developers)或使用命令行安装(适合开发者)

CLI Install命令行安装

dsh plugin --profile web add github:fashionmascherine-svg/formalswarm

把 fashionmascherine-svg/formalswarm 加入你的 DSH 配置(web profile)即可启用。

READMEREADME

FormalSwarm icon

FormalSwarm

FormalSwarm cover — thesis, antithesis, seal: a verdict you can recompute

Don't settle for an AI answer. Make it survive a debate — and a check.

FormalSwarm helps agents propose solutions, challenge assumptions, and verify the claims that can actually be tested. Start with a codebase or a folder of documents: a change to review, a funnel to design, or a plan to stress-test.

Independent agents write theses, adversarial critics challenge them, and a seal runs real commands. A deterministic rollup computes CONFIRM, REVISE, or INCONCLUSIVE from reported checks, exit codes and case counts — not a confidence score. The verdict covers the checks you defined, not every recommendation the agents make.

Built for repositories, also usable for guided, document-driven problem solving. From a 6-agent pilot to hundreds of agents, on DeepSeek Harness, Claude Code and ZCode from one and the same code.

ci license node runtimes agent calls to validate discussions

Start with code · Start with documents · Install · Check the proof

More agents can give you more ideas. FormalSwarm gives those ideas an adversary — and gives testable claims a check. If that is how you want agents to work, star FormalSwarm.

Showing the opening section of the README — the full document lives in the repository以上为 README 开头摘要,完整文档在仓库内 · View the full README on GitHub →在 GitHub 查看完整 README →

← 上一个 Prev dsh-dice-game 下一个 Next dsh-plugin-commandcode-provider →