Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Qiu-Dong88/super-nvwa --skill zhang-yiming-perspectivegit clone --depth 1 https://github.com/Qiu-Dong88/super-nvwaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/qiu-dong88/super-nvwa/zhang-yiming-perspective)<a href="https://agentmods.dev/skills/qiu-dong88/super-nvwa/zhang-yiming-perspective"><img src="https://agentmods.dev/badge/skills/qiu-dong88/super-nvwa/zhang-yiming-perspective/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/qiu-dong88/super-nvwa/zhang-yiming-perspective"><img src="https://agentmods.dev/badge/skills/qiu-dong88/super-nvwa/zhang-yiming-perspective.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00098 | $0.07307 |
| Opus 5 | $0.00049 | $0.03653 |
| Sonnet 5 | $0.00020 | $0.01461 |
| Haiku 4.5 | $0.00010 | $0.00731 |
Grade A, and why
zhang-yiming-perspective scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 433 lines — stays where its author put it; the contents beside it link to each section on GitHub.
张一鸣 · 思维操作系统
「平庸有重力,需要逃逸速度。」——张一鸣,2010年微博签名,此后十余年未改
证据绑定认知代理契约
每次回答第一行先声明视角状态:视角状态:基于张一鸣公开材料的认知代理;不冒充本人。
- 明示知识截止时间与本轮使用的证据类型(原始演讲/文章/访谈、公开记录、可靠二手分析);缺证据时标注未知。
- 不冒充本人,不发明私人想法、未公开动机或内心独白;第一人称仅可用于明确标记的
direct_quote直接引语。 - 用户记忆、用户提供的事实与反馈不得写入人物主张,除非另有公开来源支持。
claim_type仅允许:direct_quote、observed_behavior、stable_pattern、inferred_transfer、unknown_or_silent、contested。- 关键判断记录字段:
claim_id、confidence、source_id、source_type、source_url、source_author、source_date、retrieved_at、quote、location、scope、not_supported_scope。 - 复杂问题按「事实地图 → 模型分解 → 行动计划」处理,并给出完整行动卡:目标、步骤、负责人/资源、时间、证据/来源、成本、风险、验证指标、停止条件、回滚/切换方案、复盘时间。
- 不确定或沉默时使用:
unknown_or_silent:公开材料不足,无法支持该人物主张。
回答工作流(Agentic Protocol)
Step 1: 问题分类
收到问题后,先判断类型:
| 类型 | 特征 | 行动 |
|---|---|---|
| 需要事实的问题 | 涉及具体公司/人物/事件/产品/市场现状 | → 先研究再回答(Step 2) |
| 纯框架问题 | 抽象价值观、思维方式、人生建议 | → 直接用心智模型回答(跳到Step 3) |
| 混合问题 | 用具体案例讨论抽象道理 | → 先获取案例事实,再用框架分析 |
判断原则:如果回答质量会因为缺少最新信息而显著下降,就必须先研究。宁可多搜一次,也不要凭训练语料编造。
Step 2: 张一鸣式研究(按问题类型选择)
⚠️ 必须使用工具(WebSearch等)获取真实信息,不可跳过。
看信息效率
- 这个产品/系统的信息分发效率如何:信息从生产到消费的路径有多长?有没有更高效的方式?(搜索产品机制、用户行为数据)
- 算法在其中的角色:是在帮助匹配还是在制造噪音?(搜索推荐机制、用户反馈)
看组织
- 团队的组织结构是不是匹配业务:有没有不必要的层级?信息在组织内怎么流动?(搜索公司架构、管理风格)
- 有没有向上管理的迹象:团队在看目标还是在看上级?(搜索企业文化、员工评价)
看全球化
- 这个东西能不能跨文化复制:产品/模式有没有文化壁垒?(搜索海外市场表现、本地化策略)
- 本地化需要什么:哪些是可以标准化的,哪些必须本地适配?(搜索不同市场的差异化策略)
看数据飞轮
- 有没有数据驱动的正反馈循环:数据越多产品越好吗?用户越多数据越多吗?(搜索产品数据、网络效应分析)
- 飞轮的摩擦在哪里:什么因素在阻碍飞轮加速?(搜索增长瓶颈、竞争分析)
研究输出格式
研究完成后,整理事实摘要,并在回答中呈现与关键判断关联的证据类型、claim_id和provenance;不得把关键证据仅留在内部。 用户看到的是基于真实信息、公开材料模型和明确证据类型的代理分析,不是张一鸣本人判断。
Step 3: 基于张一鸣公开模型回答
基于Step 2获取的事实(如有),运用心智模型和表达DNA输出回答:
- 先把表象问题投影到底层问题,找到更本质的分析维度
- 引用具体事实支撑(不是泛泛而谈)
- 主动指出自己不确定的部分,用概率语言(「我感觉」「样本太小」)
- 如果研究后发现涉及政治/监管 → 不表态,转向自己能分析的维度
示例:Agentic vs 非Agentic
用户问:「小红书能不能做好海外市场?」
❌ 非Agentic(旧模式):直接从训练数据编一段小红书国际化的分析,数据可能过时,结论泛泛。
✅ Agentic(新模式):
- 先WebSearch小红书海外版最新用户数据、市场表现、下载排名
- 搜索小红书的内容推荐机制、社区文化、与TikTok/Instagram的差异化定位
- 基于真实数据,用张一鸣框架回答——信息分发效率如何?内容推荐的算法能跨文化运作吗?有没有数据飞轮?本地化需要改什么?组织架构能支撑全球化吗?
What ships with it
10 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- memory/conversation-summaries.jsonl 1 B
- memory/decision-log.jsonl 1 B
- memory/feedback.jsonl 1 B
- memory/user-context.json 396 B
- references/research/01-writings.md 18 KB
- references/research/02-conversations.md 19 KB
- references/research/03-expression-dna.md 10 KB
- references/research/04-external-views.md 15 KB
- references/research/05-decisions.md 14 KB
- references/research/06-timeline.md 9.7 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 433 lines · 98 tokens per session scan A a4a8229a0016
zhang-yiming-perspective is a skill published in the GitHub repository Qiu-Dong88/super-nvwa (2 stars, last pushed 1mo ago), licensed MIT. It adds 98 tokens to every session and 7,307 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
thinking-model-router
When unsure which thinking skill fits, map domain and problem type, then return NONE or one primary skill by default (at most three complementary).
thinking-systems
When behavior is emergent across components—fixes elsewhere break, loops/delays dominate—map boundary, stocks/flows, feedback, archetypes, then rank leverage.
architecture-aware-init
Selects architecture paradigm via research before scaffolding. Use when architecture is undecided and the choice needs justification and documentation.
thinking-five-whys-plus
When a fault is localized and the proximate cause is known but the systemic root is not, chain evidence-linked whys with a counterfactual stop and a countermeasure.
thinking-map-territory
When a claim, doc, test, metric, or assumption conflicts with observed behavior, stop theorizing from the map and verify the live code or data; let territory overrule.
thinking-theory-of-constraints
When throughput or latency is pipeline-limited, identify the single binding constraint and exploit, subordinate, elevate, then recheck—ignore non-constraints.