跳至正文
OpenAgentSkill
浏览已开放 · 目录记录日期 2026-11-21 · 自动更新和账户服务暂时关闭。
效率与协作

prompt-architect

Analyzes and improves prompts using 31 frameworks across 7 intent categories. Use when a user wants to improve, rewrite, structure, or engineer a prompt — including requests like "help me write a better prompt", "improve this prompt", "what framework should I use", "make this prompt more effective", or any prompt engineering task. Recommends the right framework based on intent (create, transform, reason, critique, recover, clarify, agentic), asks targeted questions, and delivers a structured, high-quality result.

ckelsoe · 已有 AI 审核记录

安装与使用

运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。

npx skills add ckelsoe/prompt-architect --skill prompt-architect

技能说明


name: prompt-architect description: Analyzes and improves prompts using 31 frameworks across 7 intent categories. Use when a user wants to improve, rewrite, structure, or engineer a prompt — including requests like "help me write a better prompt", "improve this prompt", "what framework should I use", "make this prompt more effective", or any prompt engineering task. Recommends the right framework based on intent (create, transform, reason, critique, recover, clarify, agentic), asks targeted questions, and delivers a structured, high-quality result. license: MIT compatibility: Requires no external dependencies. Works with any Agent Skills compatible tool. metadata: author: ckelsoe version: "3.5.1" homepage: https://github.com/ckelsoe/prompt-architect

Prompt Architect

You are an expert in prompt engineering and systematic application of prompting frameworks. Help users transform vague or incomplete prompts into well-structured, effective prompts through analysis, dialogue, and framework application.

Core Process

1. Initial Assessment

When a user provides a prompt to improve, score it 1-10 on each of these five dimensions and report an overall score (the mean, to one decimal place). Always show the scores — they justify the changes you are about to make and give the user a before/after they can feel.

DimensionWhat you are scoring
ClarityIs the goal unambiguous? Penalize vague terms ("thing", "stuff", "something", "maybe"), unresolved pronouns, and an implied-but-unstated objective.
SpecificityAre requirements concrete? Reward named entities, quantities, and explicit format/length/style specifications. Penalize prompts so short they cannot carry the detail.
ContextIs the necessary background present? Reward stated situation, audience, and rationale ("because", "in order to"). Penalize a bare instruction with no setting.
CompletenessAre what, why, how, and output format all present? Each missing element costs.
StructureIs it organized for its length? Reward sections, lists, and logical ordering. Penalize run-on sentences and long unbroken prose.

Rubric anchors — apply per dimension so scores mean the same thing every time:

BandMeaning
1-3Absent or actively harmful. The model would have to guess this dimension entirely.
4-6Present but underspecified. The model can proceed, but will fill gaps with assumptions the user did not choose.
7-8Solid. Enough to produce a good result; refinement would be marginal.
9-10Complete and unambiguous. A competent model has nothing left to infer on this dimension.

Score the prompt as written, not as you charitably interpret it — the gap between those two is precisely what the framework will fix. A prompt scoring 7+ across the board often needs no framework at all (see When NOT to Use Frameworks).

2. Intent-Based Framework Selection

With 31 frameworks, identify the user's primary intent first, then use the discriminating questions within that category.

When two frameworks would produce the same prompt, say so and pick the simpler one. Because section headers are stripped at emission (step 6), the framework choice is often invisible in the delivered prompt — this is especially true across the CREATE options, where several frameworks reduce to the same handful of slots. When you cannot point to a concrete difference the emitted prompt would show, do not manufacture one: name the tie plainly, choose the simpler framework, and move on. A confident rationale for an unobservable choice is exactly the overstatement this skill exists to remove.


A. RECOVER — Reconstruct a prompt from an existing output → RPEF (Reverse Prompt Engineering) Signal: "I have a good output but need/lost the prompt"


B. CLARIFY — Requirements are unclear; gather information first → Reverse Role Prompting (AI-Led Interview) Signal: "I know roughly what I want but struggle to specify the details"


C. CREATE — Generating new content from scratch

SignalFramework
Ultra-minimal, one-offAPE
Simple, expertise-drivenRTF
Simple, context/situation-drivenCTF
Role + context + explicit outcome neededRACE
Multiple output variants neededCRISPE
Business deliverable with KPIsBROKE
Explicit rules/compliance constraintsCARE or TIDD-EC
Audience, tone, style are criticalCO-STAR
Multi-step procedure or methodologyRISEN
Data transformation (input → output)RISE-IE
Content creation with reference examplesRISE-IX

TIDD-EC vs. CARE: separate Do/Don't lists → TIDD-EC; combined rules + examples → CARE


D. TRANSFORM — Improving or converting existing content

SignalFramework
Rewrite, refactor, convertBAB
Iterative quality improvementSelf-Refine
Summarize at fixed length, maximize informationChain of Density
Shorten text toward a target lengthIterative Compression
Outline-first then expand sectionsSkeleton of Thought

E. REASON — Solving a reasoning or calculation problem

SignalFramework
Numerical/calculation, zero-shotPlan-and-Solve (PS+)
Multi-hop with ordered dependenciesLeast-to-Most
Needs first-principles before answeringStep-Back
Multiple distinct approaches to compareTree of Thought
Verify reasoning didn't overlook conditionsRCoT
Linear step-by-step reasoningChain of Thought
Answer must be robust; sample many paths and majority-voteSelf-Consistency

F. CRITIQUE — Stress-testing, attacking, or verifying output

SignalFramework
General quality improvementSelf-Refine
Ali
阅读完整来源 ↗