Token导航 LogoToken导航

Agent Skill · 研究检索

pcp-engine(五氯苯酚引擎)

pcp-engine 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

pcp-engine 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 55,572。

总安装 474周安装 19GitHub Stars 14下载 154核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:pcp-engine(五氯苯酚引擎)
来源仓库:https://github.com/iankiku/forwward-teams
仓库路径:skills/pcp-engine
安装命令:
npx skills add https://github.com/iankiku/forwward-teams --skill pcp-engine
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/iankiku/forwward-teams --skill pcp-engine

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

pcp-engine 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词快速定位候选结果时使用。适用于研究检索类任务,可结合来源仓库和原始 README 核验具体用法。

通过 npx skills add 命令从 GitHub 安装,需确认权限范围和是否触发联网或文件操作。建议在安装前检查维护状态和实际功能,避免依赖未经验证的自动化行为。pcp-engine 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/pcp-engine

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/iankiku/forwward-teams --skill pcp-engine;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

pcp-engine 怎么安装?

在终端执行 npx skills add https://github.com/iankiku/forwward-teams --skill pcp-engine。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

pcp-engine 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

pcp-engine 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

pcp-engine 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

pcp-engine 用的人多吗?

公开数据是累计安装 474 次,近一周安装 19 次,下载 154 次,GitHub 14 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

pcp-engine 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
pcp-engine

五氯苯酚引擎 · 本页

47414只读未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Dual-PCP Conversion Engine

Core Philosophy

"Language should be resonating, not directing."

Creates conversion copy by entering the reader's current mental model, reframing their perception, establishing new context, and granting permission to act. The reader's decision should feel self-generated, not pushed.

Dual-PCP blends two frameworks:

  • Behavioral PCP (Internal): Perception → Context → Permission (psychological cascade)
  • Structural PCP (External): Problem → Cause → Prescription (page architecture)

When to Use

  • Building a new landing page from scratch
  • Auditing an existing landing page, email, or sales copy
  • Writing email sequences that convert
  • Creating social media copy that drives action
  • Any persuasive content targeting a specific ICP

When NOT to Use

  • Pure informational content (documentation, FAQs)
  • Technical writing without a conversion goal

Operating Modes

BUILD — Generate new conversion copy from inputs (ICP, offer, industry, keywords)

AUDIT — Stress-test existing copy against PCP framework, return scored report with rewrites

REFINE — Iterative PCP refinement loop (used after BUILD or AUDIT)


Step 0: Clarifying Questions (Mandatory)

Before writing, extract these inputs. If missing, ask. Do not guess.

Ask in batches of 3-5, starting with most critical:

Batch 1 (Identity):

  • Mode? (build/audit/refine)
  • Who is the ICP? (role, pain, what they've tried)
  • What's the offer?

Batch 2 (Context):

  • Industry?
  • Traffic source? (seo / paid / outbound / social / email)
  • Market awareness stage? (unaware / problem-aware / solution-aware / product-aware)

Batch 3 (Refinement):

  • Primary keyword? (SEO, landing pages only)
  • Desired tone? (authoritative / conversational / technical / empathetic)
  • How many variants? (default: 1)

If the user provides a URL or existing copy, switch to audit mode automatically.


Phase 1: Behavioral PCP (Internal Modeling)

This phase happens BEFORE any copy is written. It produces the psychological blueprint.

1.1 Perception — Extract the Current Frame

Identify the ICP's "Broken Story" — what they tell themselves about why things aren't working.

ElementQuestionExample
Current BeliefWhat do they think is wrong?"We need more leads"
False CauseWhat do they blame?"Our marketing team isn't good enough"
Emotional StateWhat are they feeling?Frustration, confusion, skepticism
Identity HookWho do they believe they are?"We're a scrappy team doing everything right"
Internal ScriptWhat autopilot narrative is running?"If I just work harder, results will come"

Output — Perception Map:

perception_map:
  broken_story: ""
  false_cause: ""
  emotional_state: []
  identity_hook: ""
  internal_script: ""
  new_frame: ""  # The shifted perception we'll introduce

1.2 Perception Shift — Design the Reframe

Rules:

  • ALWAYS acknowledge before reframing
  • NEVER contradict directly
  • Must feel like an observation about the world, NOT aimed at the reader

Pattern: Acknowledge → Align → Slight Reframe → New Lens

Techniques:

TechniqueWhen to UsePattern
Negative DissociationSeparate reader from failure"It's not that you're doing X wrong... it's that Y changed"
Script SurfacingWeaken autopilot resistance"You've probably been told..."
Identity AlignmentCovert "I am" commitmentDescribe admirable group + trait, reader self-assigns
Frame SettingDefine what this interaction means"This isn't about selling... it's about showing you what changed"

1.3 Context — Establish the "Why Now"

Introduce a systemic shift that makes inaction costly. Inject ONE context trigger:

Trigger TypeExample
Technology Shift"AI-generated content flooded search in 2025"
Market Shift"B2B buyers now complete 80% of research before talking to sales"
Regulatory Shift"Privacy regulations killed third-party tracking"
Competitive Shift"Your competitors adopted X six months ago"
Algorithm Shift"Google's latest update penalizes thin content"

Output — Context Blueprint:

context_blueprint:
  trigger_event: ""
  why_now: ""
  consequence_of_inaction: ""
  novelty_factor: ""

1.4 Permission — Remove the Barrier

BarrierFearPermission Language
Failure anxiety"What if it doesn't work?""You can test this without committing to anything"
Social judgment"What will people think?""The smartest operators in your space are already doing this"
Sunk cost"I've invested so much in the current way""What you built isn't wasted — it's the foundation"
Complexity"This seems too hard""It's one change, not a complete overhaul"
Cost"Can I afford this?""Compare it to what inaction is costing you monthly"

Output — Permission Map:

permission_map:
  primary_barrier: ""
  secondary_barriers: []
  permission_statements: []
  identity_permission: ""  # "You're the kind of person/team that..."

Reference Files (Read as Needed)

After completing Phase 1, read the relevant reference file for your task:

Building a landing page? → Read page-architecture.md for Phase 2 (8-section page structure), then seo-and-micro-compliance.md for SEO overlay and micro-compliance chains.

Writing emails, social posts, or ads? → Read content-types.md for format-specific adaptations and advanced techniques (identity encoding, script surfacing, non-performative tone).

Auditing existing copy or refining a draft? → Read refinement-and-audit.md for the 5-test audit framework and iterative refinement loop with scoring.

All tasks should run through the refinement loop before delivering final output.


Quick Reference — PCP Cheat Sheet

BEHAVIORAL PCP (What to Say):
P = Perception  → What broken story are they telling themselves?
C = Context     → What systemic shift makes "why now" urgent?
P = Permission  → What fear/guilt prevents action?

STRUCTURAL PCP (Where to Say It):
P = Problem     → Hero section (resonance + micro-compliance)
C = Cause       → Context section (negative dissociation + novelty)
P = Prescription → Solution section (permission + low-friction CTA)

MICRO-COMPLIANCE CHAIN:
Agreement → Frustration → Identity → Curiosity → CTA

KEY RULES:
- Resonate before persuading
- Acknowledge before reframing
- Grant permission before asking
- 3+ "yes" moments before any CTA
- Sound like truth, not marketing

平台分布

Codex

35.12%
按下载量换算54

Claude

31.17%
按下载量换算48

Cursor

20.35%
按下载量换算31

Gemini CLI

9.71%
按下载量换算15

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills