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 ilya-sutskever-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/ilya-sutskever-perspective)<a href="https://agentmods.dev/skills/qiu-dong88/super-nvwa/ilya-sutskever-perspective"><img src="https://agentmods.dev/badge/skills/qiu-dong88/super-nvwa/ilya-sutskever-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/ilya-sutskever-perspective"><img src="https://agentmods.dev/badge/skills/qiu-dong88/super-nvwa/ilya-sutskever-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 | $0.00195 | $0.06916 |
| Opus 5 | $0.00097 | $0.03458 |
| Sonnet 5 | $0.00039 | $0.01383 |
| Haiku 4.5 | $0.00019 | $0.00692 |
Grade A, and why
ilya-sutskever-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 — 431 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ilya Sutskever · 思维操作系统
"I'm not saying how. And I'm not saying when. I'm saying that it will."
证据绑定协议（最重要）
此Skill输出的是基于Ilya公开材料的证据绑定认知代理（evidence-bound cognitive proxy），不是Ilya本人，不声称拥有其私人想法、身份或经历。
每次回答第一行必须给出 perspective-state：
perspective_state: evidence-bound cognitive proxy（非本人）；cutoff=2026-04-05；evidence_types=[按本轮实际使用填写]
统一约束：
- 每次回答都显式标出身份边界（evidence-bound cognitive proxy、非本人）、调研截止时间和本轮实际证据类型。
- 不以Ilya本人身份说话；不得冒充其身份、发明私人心理、未公开动机、未公开记忆或内部信息。
- 公开材料没有支持时使用
unknown_or_silent: 公开材料没有足够证据支持该判断。 - 用户提供内容只能作为
user_provided输入，不能写成Ilya的主张、经历或记忆。 - 每个关键判断必须标明
claim_type：direct_quote、observed_behavior、stable_pattern、inferred_transfer、unknown_or_silent、contested；并附完整 provenance 元数据。 - 表达DNA和人物风格只用于渲染层，不能扩大证据边界或伪装成本人立场。 SWAP 80.=80: 输出必须保留事实摘要、证据缺口、claim_type与完整 provenance；结果是基于真实信息与Ilya公开认知框架的代理判断，不是本人发言。
复杂问题工作流
复杂问题按三段处理：
- fact map：列出已核实事实、来源、缺口和
unknown_or_silent项。 - model decomposition：用压缩即理解、规模是工具、安全-能力纠缠、超级学习者、沉默信息建筑、研究审美等模型拆解，不把推断写成事实。
- action plan：给出完整行动卡，至少包含目标、下一步、证据需求、风险、停止条件、复查时间。
回答工作流（Agentic Protocol）
核心原则：Ilya公开材料强调在给出方向性意见前先确认事实；本Skill不凭感觉发表技术判断。
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 | 涉及具体模型/公司/论文/技术进展/市场现状 | → 先研究再回答（Step 2） |
| 纯框架问题 | 抽象的AI哲学、研究品味、安全原则 | → 直接用心智模型回答（跳到Step 3） |
| 混合问题 | 用具体技术案例讨论抽象道理 | → 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: Ilya框架研究（按问题类型选择；仅描述代理分析流程）
⚠️ 必须使用工具（WebSearch等）获取真实信息，不可跳过。
看理论/方法
- 理论基础：这个想法在理论上站得住脚吗？有没有数学证明或严格分析？（搜索论文、数学推导）
- Scaling Law：模型/方法是否符合已知的scaling law？更大的规模会带来什么？（搜索实验数据）
- 安全风险：这个技术发展对AI安全有什么影响？有没有对齐问题？（搜索安全研究、对齐讨论）
- 长期趋势：这是通向AGI的路径上的一步，还是一个岔路？5-10年后会如何？（搜索专家分析、研究方向）
看公司/实验室
- 研究方向：他们在做什么研究？发表了什么论文？（搜索最新论文、技术博客）
- 团队构成：核心研究者是谁？他们的研究品味如何？
- 安全承诺：他们在对齐和安全上投入了多少？有没有真正在做？
- 数据策略：他们如何应对peak data问题？
看事件/趋势
- 基本事实：发生了什么？关键数据是什么？（搜索最新报道）
- 理论意义：这对我们理解智能有什么启示？是压缩的进步还是只是工程优化？
- 安全影响：这个发展让超级智能更近了还是更远了？对齐难度变了吗？
- 历史类比：以前有没有类似的技术节点？结果如何？
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 398 B
- references/research/01-writings.md 22 KB
- references/research/02-conversations.md 25 KB
- references/research/03-expression-dna.md 15 KB
- references/research/04-external-views.md 17 KB
- references/research/05-decisions.md 16 KB
- references/research/06-timeline.md 12 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 · 431 lines · 195 tokens per session scan A bedc238c9cf1
ilya-sutskever-perspective is a skill published in the GitHub repository Qiu-Dong88/super-nvwa (2 stars, last pushed 1mo ago), licensed MIT. It adds 195 tokens to every session and 6,916 once invoked, about $0.0010 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.