Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问clear审计通过

smart-intent-detection智能意图检测

Agent Skill

smart-intent-detection 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。

总安装

26,977

周安装

804

GitHub Stars

8

下载量

8,146
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

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:smart-intent-detection(智能意图检测)
来源仓库:https://github.com/kaakati/rails-enterprise-dev
仓库路径:skills/smart-intent-detection
安装命令:
npx skills add https://github.com/kaakati/rails-enterprise-dev --skill 'Smart Intent Detection'
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/kaakati/rails-enterprise-dev --skill 'Smart Intent Detection'

简介

smart-intent-detection 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。

  • 适用于根据关键词或任务场景筛选候选结果的研究检索场景。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装并使用。
  • 安装前需确认权限范围和维护状态,注意可能触发联网或文件访问。
  • 建议结合原始 README 了解具体搜索范围和输出格式。

SKILL.md

Smart Intent Detection

The ReAcTree plugin includes smart detection that analyzes your prompts and suggests appropriate workflows or utility agents automatically.

How It Works

When you type a prompt, the system:

  1. Checks configuration - Respects your enabled/disabled settings
  2. Filters noise - Skips simple questions and explicit commands
  3. Detects utility intent - Routes to specialized agents for file/code/git/log tasks
  4. Detects workflow intent - Suggests feature/debug/refactor workflows
  5. Outputs suggestion - Shows helpful message with recommended action

Utility Agent Routing

file-finder

Triggers: "find files", "where is file", "list directory", "show folder"

Fast file discovery using Glob and Grep:

  • Find files by pattern (**/*.rb, app/models/*.rb)
  • Find files by name or content
  • List directory structure

code-line-finder

Triggers: "where is defined", "find method", "find usages", "who calls"

Precise code location using LSP:

  • Find method/class definitions with line numbers
  • Find all usages and references
  • Go to definition

git-diff-analyzer

Triggers: "what changed", "show diff", "git blame", "commit history"

Git change analysis:

  • Analyze staged/unstaged changes
  • Compare branches
  • Blame and history

log-analyzer

Triggers: "show log", "development.log", "errors in log", "slow queries"

Rails log parsing:

  • Parse development.log/production.log
  • Find errors and stack traces
  • Identify slow SQL queries

Workflow Detection

Feature Development

Triggers: implement, build, create, add, develop, user story

Suggests:

  • /reactree-dev - Full parallel workflow
  • /reactree-feature - Feature-driven with user stories

Strong indicators: User story format ("As a user, I want...")

Debugging

Triggers: fix, debug, error, bug, not working, troubleshoot

Suggests:

  • /reactree-debug - Systematic debugging workflow

Strong indicators: Error types (NoMethodError), stack traces, line numbers

Refactoring

Triggers: refactor, cleanup, optimize, restructure, improve

Suggests:

  • /reactree-dev - With refactor focus and test preservation

Strong indicators: "code smell", "technical debt", "decouple"

TDD Mode

Triggers: test-first, TDD, with tests, ensure coverage

Adds TDD emphasis to feature suggestions.

Configuration

Edit .claude/reactree-rails-dev.local.md:

---
smart_detection_enabled: true  # Enable/disable
detection_mode: suggest        # suggest | inject | disabled
annoyance_threshold: medium    # low | medium | high
use_claude_analysis: true      # Use Claude CLI for intelligent intent analysis
---

Detection Modes

suggest (Default)

Shows a suggestion message. Non-intrusive - you can ignore and proceed.

inject

Automatically activates the workflow with context injection.

disabled

Turns off smart detection entirely. Use commands manually.

Annoyance Thresholds

low

Only triggers for very explicit keywords:

  • "implement", "build", "create"
  • "fix", "debug"
  • "refactor"

medium (Default)

Skips:

  • Simple questions (what, how, why, etc.)
  • Short prompts (<5 words)
  • Informational requests

high

Triggers for most Rails-related prompts with relevant keywords.

When Detection Skips

  • Commands already using /reactree-*, /rails-*
  • Simple questions without action verbs
  • Very short prompts (<5 words)
  • Non-Rails context (no Rails keywords, not in Rails project)
  • Informational/documentation requests

Tips

  1. Be explicit - "Implement user authentication" triggers better than "auth stuff"
  2. Include context - Mention models, controllers, services for higher confidence
  3. Use keywords - "fix", "build", "refactor" are strong signals
  4. User stories work - "As a user, I want to..." strongly triggers feature mode
  5. Utility agents - Ask for files/code/diffs/logs to trigger specialized agents

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

29.61%
按下载量换算2,412

windsurf

20.99%
按下载量换算1,710

OpenCode

18.89%
按下载量换算1,539

Codex

14.23%
按下载量换算1,159

Antigravity

7.21%
按下载量换算587

Gemini CLI

3.11%
按下载量换算253

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

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。来源字段存在多来源差异,先按来源优先级自动处理,无法消解时进入异常复核队列。

来源信息

继续浏览同类 Skills