formalswarm
运行「正题—反题—封印」三方辩论，其结论由真实命令的退出码与用例数计算得出，绝不采信智能体的文字陈述。
Runs a thesis-antithesis-seal debate whose verdict is computed from real command exit codes and case counts, never from agent prose.
怎么安装
dsh plugin add github:fashionmascherine-svg/formalswarm git clone https://github.com/fashionmascherine-svg/formalswarm ~/.claude/skills/formalswarm 简介
FormalSwarm **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…
推荐参考
信息
- 协议
- MIT
- 语言
- JavaScript
- GitHub 星标
- 2
- 月下载
- –
- 最近更新
- 2026-09-17
- 创建于
- 2026-09-15
基础安全检查
- 检查结果
- curated 收录但无 npm 包/安装命令
- 收录来源
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- 主题标签
- adversarial, agentic, ai-agents, claude-code, code-review, deepseek, deepseek-harness, developer-tools, dsh-plugin, dsh-plugin-market, dsh-plugins, llm, multi-agent, testing, verification, zcode
同类推荐
SuperClaude_Framework
SuperClaude-Org/SuperClaude_Framework
一个配置框架，通过专用命令、认知角色和开发方法论增强Claude Code。