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 devcxl/mattpocock-skills-zh --skill ask-mattgit clone --depth 1 https://github.com/devcxl/mattpocock-skills-zhWrote 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/devcxl/mattpocock-skills-zh/ask-matt)<a href="https://agentmods.dev/skills/devcxl/mattpocock-skills-zh/ask-matt"><img src="https://agentmods.dev/badge/skills/devcxl/mattpocock-skills-zh/ask-matt/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/devcxl/mattpocock-skills-zh/ask-matt"><img src="https://agentmods.dev/badge/skills/devcxl/mattpocock-skills-zh/ask-matt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00024 | $0.03145 |
| Opus 5 | $0.00012 | $0.01572 |
| Sonnet 5 | $0.00005 | $0.00629 |
| Haiku 4.5 | $0.00002 | $0.00314 |
Grade A, and why
ask-matt 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ask Matt
你不必记住每一个技能，所以来问。
流程（flow）是穿过技能的一条路径。大多数路径沿一条主干流程行进，两条入口匝道汇入其中。其余的都是独立技能，或者是在底层运行的词汇层。
主干流程：想法 → 交付
大多数工作走的路线。你有一个想法，想把它建出来。
-
/grill-with-docs— 通过盘问打磨想法。只要你在工作目录里就从这里开始：它是有状态的，会把学到的东西保存在CONTEXT.md和 ADR 中。（没有工作目录？用/grill-me——见"独立技能"。两者跑的是同一个/grilling原语；grill-with-docs是会留下书面痕迹的那个，所以只要有仓库可以留痕迹，它就是更好的选择。） -
分支——所有问题都能在对话中解决吗？ 如果某个问题需要一个可运行的答案（状态、业务逻辑、必须亲眼看到的 UI），就绕道原型，由
/handoff双向桥接（原型住在自己的目录里，这正是/handoff的用武之地——见"阶段边界"）：/handoff跳出去，然后针对那个文件开一个新会话，/prototype用一次性代码回答问题，/handoff把你学到的东西带回来，并在原始想法线程中引用它。
-
分支——这是多会话构建吗？
- 是 →
/to-spec（把对话变成 spec），然后/to-tickets把它拆成示踪子弹式 ticket，每个都声明自己的阻塞边。在本地跟踪器上，每个 ticket 是.scratch/<feature>/issues/下的一个文件，按阻塞优先的顺序人工推进；在真实跟踪器上，阻塞边变成原生阻塞链接，所以任何阻塞项已完成的 ticket 都可以被领取——每个 ticket 启动一次/implement，每个之间用/clear清空上下文。每个 ticket 自包含，所以上一个的上下文可以丢弃。 - 否 → 就在同一个上下文窗口里
/implement。
无论哪种方式，
/implement构建每个 issue 的方式都是在内部驱动/tdd——一次一个红-绿切片——然后在提交前用/code-review收尾，对 diff 做双轴审查（标准 + 规格）。只想在没有完整 spec 的情况下测试优先地构建一个具体行为时，单独用/tdd；任何时候想对照固定点审查分支或 PR，单独用/code-review。 - 是 →
上下文卫生
把步骤 1–3 保持在一个不间断的上下文窗口里——在 /to-tickets 之前不要压缩或清空——这样盘问、spec 和 ticket 都建立在同一个思考之上。之后每次 /implement 从 ticket 出发、重新开始。
这里的上限是**智能区**：模型仍能敏锐推理的窗口（最先进模型约 150k token）。如果会话在 /to-tickets 之前就逼近它，不要在劣化的状态下硬撑——在最近的阶段边界 /compact 然后继续（见"阶段边界"）。
入口匝道
一个会产生工作的起点场景，然后并入主干流程。
-
bug 和请求堆积 →
/triage。它让 issue 走过分类角色，产出 agent 就绪的 issue，之后由/implement接手。分类只适用于不是你创建的 issue——bug 报告、外来的功能请求、任何以原始形态到达的东西。
/to-tickets产出的 ticket 已经是 agent 就绪的，所以不要分类它们。 -
有东西坏了 →
/diagnosing-bugs。用于那些难啃的：一眼看不出的 bug、间歇性偶发（flake）、在两个已知正常状态之间悄悄溜进来的回归。在拥有紧反馈循环之前它拒绝空谈理论——一条已经能对这个 bug 变红的命令——然后用回归测试修复。当真正的发现是没有好的接缝来锁死 bug 时，它的事后总结会交接给/improve-codebase-architecture。 -
一项庞大而模糊的工作——绿地项目或大型功能构建，大到一次会话装不下 →
/wayfinder，这是这里认知负荷最高的流程。当从这里到目的地的路还看不见时，它在 issue 跟踪器上绘制一张由决策 ticket 组成的共享地图，一次解决一个——产出决策，而非交付物——直到迷雾被推后、道路清晰。/grill-with-docs打磨的是你能在一次会话里 hold 住的想法，wayfinder 则是为你 hold 不住的想法准备的——它更慢、更密，所以只留给这种情况，绝不要用于范围明确的功能。
What ships with it
2 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.
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 · 91 lines · 24 tokens per session scan A d622b59ab356
ask-matt is a skill published in the GitHub repository devcxl/mattpocock-skills-zh (343 stars, last pushed 7d ago), licensed MIT. It adds 24 tokens to every session and 3,145 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
chinese-commit-conventions
A Chinese-language guide to Conventional Commits, a format for writing consistent Git commit messages, plus related changelog, commit-checking, and commit-helper configuration.
chinese-documentation
A Chinese technical-documentation style guide covering spacing, punctuation, numbers, terminology, and links when Chinese and English appear together.
systematic-debugging
A step-by-step method for finding the underlying cause of technical problems before changing code. It covers reading errors, reproducing failures, checking recent changes, and tracing data across system components.
chinese-code-review
A Chinese-language code-review communication guide with templates and severity levels for review comments.
dispatching-parallel-agents
A workflow for assigning separate coding agents to independent tasks that do not share state or depend on one another. Each agent receives a focused scope, constraints, and an expected result.
executing-plans
A procedure for carrying out a written software implementation plan in a separate session, including reviewing the plan, completing its tasks, and checking the results at defined review points.