CtrlK
BlogDocsLog inGet started
Tessl Logo

huashu-speech-coach

演讲与分享教练。基于Patrick Winston(MIT AI教授)的How to Speak方法论,帮助准备线下培训、技术分享、B站教程视频等演讲场景。当用户提到"演讲"、"分享"、"培训"、"讲课"、"PPT演讲"、"开场"、"结尾"、"如何讲"、"演讲结构"时使用此技能。

69

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

77%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, actionable coaching reference with clear review workflows and concrete examples; its weaknesses are verbosity in rule rationales and a monolithic body that underuses the existing reference file for progressive disclosure.

Suggestions

Trim the explanatory rationale prose after each Winston rule to the rule plus one tight example, relying on Claude's intelligence for the why.

Move the detailed tables (幻灯片七宗罪, Winston之星, 时间与地点) into `references/patrick-winston-how-to-speak.md` and keep SKILL.md a lean overview that links to each by section, making progressive disclosure one level deep and well-signaled.

Add explicit cross-links from inline sections to the relevant part of the reference file so the reference is navigated, not just listed at the top.

DimensionReasoningScore

Conciseness

The ~300-line body is accurate and mostly domain-specific (Winston heuristics), but each rule is followed by explanatory rationale prose ('观众刚到场时还在适应——收手机、调整坐姿、适应你的声音...') that could be tightened; mostly efficient but padded in places.

2 / 3

Actionability

For an instruction-only skill it gives concrete, specific guidance: '等待时间:至少7秒', '字体≥40号', '告诉他们你要讲什么 → 讲给他们听 → 告诉他们你讲了什么', plus good/bad examples and checklists — copy-ready direction rather than vague description.

3 / 3

Workflow Clarity

The '如何使用这个Skill' section sequences four review scenarios each with explicit checklists (validation items), and '审查评分' adds a 5-dimension scoring framework plus '优先修改项' output, matching the checklist-driven level-3 anchor; no destructive ops require feedback loops.

3 / 3

Progressive Disclosure

The reference `references/patrick-winston-how-to-speak.md` is real and signaled one level deep at the top, but the SKILL.md body is a monolithic wall of text keeping all detailed methodology inline rather than acting as a lean overview that points into the reference for specifics.

2 / 3

Total

10

/

12

Passed

Description

90%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description with explicit what/when structure, natural trigger terms, and a distinct niche; the main weakness is the generic verb '帮助准备' rather than enumerating concrete coaching actions. Voice is appropriately third person, so no specificity penalty applies.

Suggestions

Replace the generic verb '帮助准备' with concrete actions the skill performs, e.g. '审查演讲大纲、优化开场结尾、检查Winston规则、模拟观众反馈、逐张审查幻灯片'.

Consider adding 'PPT审查' or '视频脚本' as explicit trigger phrases to match the scenarios already named.

DimensionReasoningScore

Specificity

Quotes a single generic verb '帮助准备' applied to concrete enumerated scenarios ('线下培训、技术分享、B站教程视频等演讲场景'), naming the domain and scenarios but not listing multiple distinct concrete actions as the level-3 anchor requires.

2 / 3

Completeness

Explicitly answers what ('演讲与分享教练...基于...How to Speak方法论,帮助准备...演讲场景') and when (explicit '当用户提到...时使用此技能' trigger clause), matching the level-3 anchor.

3 / 3

Trigger Term Quality

Provides strong natural-language coverage via '当用户提到"演讲"、"分享"、"培训"、"讲课"、"PPT演讲"、"开场"、"结尾"、"如何讲"、"演讲结构"时使用此技能' — terms a user would naturally say.

3 / 3

Distinctiveness Conflict Risk

Occupies a clear niche (Winston-methodology presentation coaching) with distinct, speech-specific triggers unlikely to fire for unrelated skills.

3 / 3

Total

11

/

12

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
alchaincyf/huashu-skills
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.