最高评分
社区最喜爱的高星技能，质量有保障。
small-reviewable-site-health-fix
tin-computer
Diagnose and propose one low-risk mechanical website improvement as a reviewable PR.
Project weekly brief
tin-computer
Do not include a Sources section; Tin appends the exact durable references after generation.
article-delivery
tin-computer
Adapt one approved article to an existing site without rewriting its copy.
content-diagram
tin-computer
Compose an editable Tin diagram with clear hierarchy, lanes, and return paths.
planned-content
tin-computer
Assess current coverage, then draft only useful, fact-grounded planned content.
public-article
tin-computer
Turn durable project evidence and original thinking into a rigorous, readable public article with a clear audience, thesis, narrative, citations, and limitations.
public-article-edit
tin-computer
Give a public article a fact-preserving final edit for plain, specific, human prose without changing its evidence, meaning, citations, or authorial constraints.
email-shortlist
tin-computer
Build a bounded outreach shortlist from connected Gmail and Calendar evidence.
product-code-map
tin-computer
Read the founder's repository and record which product surfaces are wired, who reaches them, and how they are gated, as the Code map section of project memory.
product-deep-dive
tin-computer
Map what a product actually does through its docs, its code map, and live use with a Tin-owned identity, and record the reconciled Feature map in project memory.
product-audit
tin-computer
Exercise every mapped feature of a product as a user with a Tin-owned identity, run cross-cutting quality checks, and write one dated audit report with ranked findings.
signup-walkthrough
tin-computer
Sign up for a product as a fresh user with a Tin-owned identity and report the journey.
research-deep-dive
tin-computer
Run a bounded, top-down investigation that tests project assumptions against current primary evidence and produces one decision-useful Markdown report.
site-health-improvement
tin-computer
Diagnose, implement, verify, and package one bounded website-health improvement as a reviewable GitHub pull request.
product-demo-video
tin-computer
Capture a live product at phone size and render a narrated 9:16 demo video.
audit-title-repair
tin-computer
Repair one evidenced missing-title or missing-description finding in verified HTML sources.
ship-commits
phase-rs
Use when shipping local commits to main via the merge queue. Creates an isolated worktree based on origin/main, cherry-picks the named commits into a fresh branch, pushes with --no-verify, opens a PR with gh pr create --fill, and enqueues with gh pr merge --auto. Outbound counterpart to pr-contribution-handler. Use when the user says "ship this", "push to main", "send through the queue", "PR this work", or has finished a chunk of work and wants it on main.
project-reference
phase-rs
Reference lookup for phase.rs — build/test/cargo/clippy commands, cargo aliases, WASM build, card-data pipeline and jq lookups, frontend (pnpm) commands, coverage report, crate/workspace architecture, engine internals, WASM bridge, AI engine, multiplayer server, frontend layers, environment variables, releasing (cargo-release), and CI. Use when you need to look up how to build, test, run, generate card data, release, or where a crate, module, command, or env var lives.
benchmark
bdfinst
Capture runtime performance metrics (Core Web Vitals, resource sizes, load times) for web pages. Compare against baselines and performance budgets. Use when the user says "benchmark", "check performance", "page speed", "web vitals", "performance regression", or "how fast is this page".
bloom-tutor
Li-Evan
Use when 用户想以一对一苏格拉底导师的方式系统学习一个课题——开一门新课、推进课题的下一篇、提交学习反馈或说「我读完了」、或整理/查看学习日志。基于 Bloom 2 Sigma 的交互式学习系统。触发词：开个文件夹学X、我想学X、帮我学X、继续、下一篇、我读完了、整理学习、查看学习日志、interactive Socratic tutoring、Bloom 2 sigma learning。
learn-crossover
Li-Evan
当用户学习或接触一个新概念/新技术/新算法/新领域时使用（尤其感到陌生或有点难时）。用「跨界原则」拿用户已掌握的知识快速撬动新知识——指出他其实已经学过的同一个东西（换了名字）、结构同构的旧知识、能解释新知识的已有知识，并点出新概念体现的跨领域元知识模式。让「学新东西」变成「发现你已经会了一半」。触发场景：学 X、接触 X、这个好难、X 是什么、帮我理解 X。
learn-deep
Li-Evan
用户学任何新概念/新技术/新理论的默认深度入口——一次性用五个视角把概念讲透并帮他选深入方向：crossover 用已会的撬动、occam 框定该学多深、graph 建知识地图、prototype 最小原型迭代、feynman 拷问检验。触发场景：我想学 X、理解 X、X 是什么、讲讲 X、搞懂 X、学一下 X、深入 X、给我讲讲 X。除非用户明确只要某一个视角（那时改用对应的单个 learn-* skill）。
learn-feynman
Li-Evan
当用户学完一个东西想自查是否真懂、或觉得「好像懂了」但不确定时使用。用「费曼学习法」让他用自己的话把概念讲出来，你扮好奇学生专挑他含糊/跳过的地方追问，把「讲不顺的模糊处」揪出来作为没真懂的漏洞，定位是缺前置知识还是没想透，判断理解是否闭环。触发场景：我学完了考考我、自查一下、我好像懂了、我讲讲你看对不对、检验我的理解、这个我真懂了吗。
learn-occam
Li-Evan
当用户纠结要不要学某个东西、学到什么程度，或在做时间/精力/项目取舍时使用。用「简易策略」先逼问要解决的既定问题，检验现有知识能否搞定，评估知识贬值速度与 ROI，用「探索 vs 应用」判断该学新的还是用现有的，给出「学 / 不学 / 只学最小够用」的结论，避免囤积会贬值的知识。触发场景：要不要学 X、值不值得深入、学到什么程度够、时间不够该学啥、该深挖还是够用就行。