普华永道：企业应统一Agent执行框架而非按场景自建
PwC looked into Anthropic's harness primitives (a loop with memory, filesystem,…
PwC looked into Anthropic's harness primitives (a loop with memory, filesystem, and bash, i.e. the Claude Code shape) and argues that enterprises should stop building a new agent system per use case and instead deploy one identical harness everywhere.
普华永道研究了Anthropic的“工具原语”（一个带记忆、文件系统和bash的循环，即Claude Code的形态），并主张企业应停止为每个用例构建新的代理系统，转而统一部署一个相同的工具框架。
This happens mostly because it collapses review, not because it runs better.
这主要是因为这样做简化了审查流程，而非运行效果更佳。
So they recommend standardize the loop across teams and let the prose in an instructions file carry the difference.
因此，他们建议跨团队标准化这个循环，并让指令文件中的文本内容来体现差异。
If every agent in a company runs the same unmodified harness and differs only in a plain-text instructions file, auditing the whole fleet becomes reading documents instead of reviewing code.
如果公司内的每个代理都运行相同的未修改工具框架，仅通过纯文本指令文件区分，那么对整个代理群的审计就变成了阅读文档，而非审查代码。
– arxiv. org/abs/2608.20622
– arxiv. org/abs/2608.20622
Title: "Applying Anthropic Primitives at Large Enterprises: Harness Paradigm for Knowledge Work"
标题：“在大规模企业中应用Anthropic原语：知识工作的工具框架范式”
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力