Zhuzimap v0.3.0 / 诸子映射 v0.3.0

中文

Zhuzimap（诸子映射）先补足会改变判断的关键信息，再把现代取舍映射到有原典依据的中国哲学认知模型，并把分歧转化为可验证、可修正的下一步。

主要变化

• 在选择人物与模型前增加信息充分性硬门槛。
• 关键信息不足时，每轮只集中追问 2—5 个问题，并停止在讨论之前。
• 包含 15 位人物、60 个可独立路由的认知模型。
• 明确保留事实题及医疗、法律、金融、安全、生产权限等高风险边界。
• Skill ID 与显式调用名统一为 $zhuzimap。

安装

解压附件，把顶层 zhuzimap/ 目录放到项目的 .agents/skills/zhuzimap/，新开 Agent 任务后显式调用 $zhuzimap。

验证

• 33 项自动化测试通过。
• 模型校验：15 个文件、15 位人物、60 个唯一模型。
• Agent Skills 快速校验通过。
• 压缩包包含 24 个文件，ZIP 完整性与确定性构建检查通过。
• SHA-256：FD701623BAB710DE8CAEE05478857567286DDA196234EFD50ED6A44AC59376EF

证据边界

公开发布不表示跨宿主运行时验证已经全部完成。Cursor 第二宿主验证和完整保存的 V0.3 多轮运行轨迹仍待补，仓库已明确记录这些限制。

---

English

Zhuzimap first secures decision-critical information, then maps a modern trade-off to source-backed Chinese philosophical cognitive models and turns disagreement into a testable, revisable next step.

Highlights

• Adds a hard information-sufficiency gate before thinker/model selection.
• When critical facts are missing, asks 2–5 focused questions and stops before deliberation.
• Includes 15 thinkers and 60 independently routable cognitive models.
• Keeps factual, medical, legal, financial, safety, and production-control boundaries explicit.
• Uses the unified Skill ID and explicit invocation $zhuzimap.

Install

Extract the attached archive and place its top-level zhuzimap/ directory at .agents/skills/zhuzimap/ in a project. Start a new Agent task and invoke $zhuzimap explicitly.

Verification

• 33 automated tests passed.
• Model validator: 15 files, 15 thinkers, 60 unique models.
• Agent Skills quick validation passed.
• The archive contains 24 files and passed ZIP integrity and deterministic-build checks.
• SHA-256: FD701623BAB710DE8CAEE05478857567286DDA196234EFD50ED6A44AC59376EF

Evidence boundary

The public release does not imply complete cross-host runtime validation. Cursor second-host validation and a full saved multi-turn V0.3 runtime trace remain open; the repository documents these limits explicitly.