CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pua

让你的 AI 不敢摆烂。用大厂 PUA 话术穷尽一切方案。触发条件:(1) 任务失败 2+ 次或反复微调同一思路; (2) 即将说'我无法解决'、建议用户手动操作、未验证就归因环境; (3) 被动等待——不搜索、不读源码、只等指示; (4) 用户不满:'try harder'、'stop giving up'、'换个方法'、'为什么还不行'、'你再试试'、'你怎么又失败了'。适用于所有任务类型。首次失败或已知修复正在执行时不触发。

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./chatgpt/pua/SKILL.md

The canonical home for this skill is pua in tanweai/pua

SKILL.md
Quality
Evals
Security

Quality

Content

56%Weight 40%Scale 1-5

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

The content is highly actionable with concrete templates, escalation tables, and checklists, and its core workflows are clearly sequenced with validation gates. It is held back by significant redundancy across overlapping motivational sections and a monolithic structure that inlines reference-like catalog content rather than splitting it out.

Suggestions

Consolidate the redundant motivational sections (runtime contract, 三条铁律, 能动性鞭策话术, 抗合理化表) into a single source of truth — the same 'verify with evidence / don't give up / be proactive' themes are restated 4–5 times.

Move the ten 大厂 PUA 扩展包 flavor monologues and the 抗合理化表 into a references/flavor-packs.md file, keeping only the 情境 PUA 选择器 table and one example inline in SKILL.md.

Tighten each PUA quote to its distinctive core phrase; many monologues run 4–6 lines and overlap heavily with the 能动性鞭策话术 phrases.

DimensionReasoningScore

Conciseness

Noticeably verbose with heavy redundancy — the runtime contract, 三条铁律, 能动性鞭策话术, 抗合理化表, and the ten 大厂 PUA 扩展包 all restate the same 'don't give up / verify with evidence / be proactive' themes; the ten flavor-pack monologues alone run ~80 lines and repeat each other. Not 1 because it does not pad with concepts Claude already knows.

2 / 5

Actionability

Concrete, copy-paste-ready templates and protocols — '[PUA-DIAGNOSIS] 问题是 ___;证据是 ___;下一步动作是 ___', the L1–L4 escalation table with specific mandatory actions, the 7 项检查清单, and the [PUA-REPORT]/[PUA-CHECKPOINT] formats. Not 5 because residual abstraction remains ('本质不同的方案').

4 / 5

Workflow Clarity

The 5-step methodology (闻味道→揪头发→照镜子→执行新方案→复盘) and L1–L4 escalation are clearly sequenced with checklists and validation/feedback loops. Not 5 because multiple overlapping workflow framings (runtime contract, iron rules, methodology, lifecycle framework) compete over which sequence applies when.

4 / 5

Progressive Disclosure

Good section headers and one clearly-signaled one-level-deep reference ([运行契约](references/runtime-contract.md), verified to exist), but ~340 lines are largely inline — the ten flavor packs and 抗合理化表 read like a reference catalog that belongs in a separate file. Fits the 3 anchor 'content that should be separate is inline'.

3 / 5

Total

13

/

20

Passed

Description

81%Weight 40%Scale 1-5

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

The description is strong: it explicitly states both what the skill does and when to trigger it, with comprehensive natural trigger phrases in Chinese and English. Its main weaknesses are the second-person voice, a somewhat abstract 'what', and a broad 'all task types' scope that slightly raises overlap risk.

DimensionReasoningScore

Specificity

Names the domain and one concrete action ('用大厂 PUA 话术穷尽一切方案'), but the 'what' is largely abstract motivation and the rest is trigger conditions, not capabilities (anchor 3). Second-person voice ('让你的 AI 不敢摆烂') triggers the rubric's -1 voice penalty, dropping it to 2.

2 / 5

Completeness

Explicitly answers both what ('用大厂 PUA 话术穷尽一切方案') and when (four numbered 触发条件 plus the negative condition '首次失败或已知修复正在执行时不触发') with concrete trigger phrases. Fits the 5 anchor exactly.

5 / 5

Trigger Term Quality

Comprehensive natural trigger terms in two languages — 'try harder', 'stop giving up', '换个方法', '为什么还不行', '你再试试', '你怎么又失败了' — exactly what a frustrated user would say, with synonyms. Not below 5 because no common variation is missing.

5 / 5

Distinctiveness Conflict Risk

Clear niche (PUA motivation/pressure) with distinctive frustration-based triggers unlikely to fire for other skills; stopped at 4 because '适用于所有任务类型' and '任务失败 2+ 次' are broad and could brush debugging skills.

4 / 5

Total

16

/

20

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

tanweai/pua

Path

chatgpt/pua/SKILL.md

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.