Co-Pilot / 辅助式
更新于 24 days ago

parallel-findall-cli

Jjanwilmake
0.0k
janwilmake/parallel-findall-cli
82
Agent 评分

💡 摘要

一个命令行工具,用于基于复杂标准发现实体,使用Parallel FindAll API。

🎯 适合人群

寻求找到特定公司的数据分析师寻找潜在客户的业务发展专业人士需要评估行业趋势的研究人员寻找潜在初创公司的投资者将实体搜索集成到应用程序中的开发人员

🤖 AI 吐槽:看起来很能打,但别让配置把人劝退。

安全分析中风险

风险:Medium。建议检查:是否执行 shell/命令行指令;是否发起外网请求(SSRF/数据外发);API Key/Token 的获取、存储与泄露风险;文件读写范围与路径穿越风险;依赖锁定与供应链风险。以最小权限运行,并在生产环境启用前审计代码与依赖。

FindAll CLI

A command-line interface for the Parallel FindAll API that helps you discover and evaluate entities matching complex criteria.

Installation

npm install -g findall-cli

Or use directly with npx:

npx findall-cli run ./output "Find AI companies that raised Series A in 2024"

Installation into your coding agent (Claude, Cursor, etc.)

npx skills add janwilmake/parallel-findall-cli

Usage

Basic Usage

findall run <output-dir> <objective>

Example:

findall run ./results "Find all SaaS companies with SOC2 Type II certification"

With Options

# Specify API key findall run ./results "..." --api-key YOUR_API_KEY # Use different generator tier findall run ./results "..." --generator pro # Set match limit findall run ./results "..." --limit 100 # Skip interactive preview findall run ./results "..." --skip-preview --auto-approve

Environment Variables

export PARALLEL_API_KEY=your_api_key_here findall run ./results "Find companies..."

Workflow

The CLI follows this workflow:

  1. Ingest: Converts your natural language objective into a structured schema
  2. Preview: Shows the generated schema and allows editing
  3. Configure: Set generator tier, match limits, and other options
  4. Run: Executes the FindAll search with real-time progress
  5. Output: Saves results to specified directory

Interactive Schema Editor

The CLI provides an interactive editor to:

  • View and modify the objective
  • Change entity type
  • Add, edit, or remove match conditions
  • Preview the full JSON schema

Use arrow keys to navigate and follow the prompts.

Output Files

Results are saved to the specified output directory:

  • summary.json - Run metadata and statistics
  • matches.json - All matched candidates with full details
  • candidates.json - All evaluated candidates (matched and unmatched)
  • schema.json - Final schema used for the run
  • matches.csv - CSV export of matched candidates

Options

| Option | Description | Default | | ------------------------ | ----------------------------------------- | ------------------- | | -k, --api-key <key> | Parallel API key | $PARALLEL_API_KEY | | -g, --generator <tier> | Generator tier (base|core|pro|preview) | core | | -l, --limit <number> | Initial match limit (5-1000) | 50 | | --skip-preview | Skip schema preview | false | | --auto-approve | Auto-approve schema without editing | false |

Examples

Find AI companies with specific criteria

findall run ./ai-companies \ "Find AI companies that raised Series A in 2024 and have at least 50 employees" \ --generator pro \ --limit 100

Find security-compliant SaaS companies

findall run ./saas-companies \ "Find SaaS companies with SOC2 Type II and ISO 27001 certifications" \ --generator core

Non-interactive mode

export PARALLEL_API_KEY=your_key findall run ./results "Your objective here" --skip-preview --auto-approve

Development

# Install dependencies npm install # Build npm run build # Run locally npm start run ./output "Test objective" # Watch mode npm run dev

API Documentation

This CLI uses the Parallel FindAll API. For detailed API documentation, see: https://api.parallel.ai/docs

License

MIT

五维分析
清晰度8/10
创新性7/10
实用性9/10
完整性9/10
可维护性8/10
优缺点分析

优点

  • 灵活的命令行界面,适用于各种查询
  • 支持多种输出格式
  • 交互式模式编辑器,便于自定义

缺点

  • 需要API密钥才能充分发挥功能
  • 新用户可能需要学习曲线
  • 仅限于Parallel FindAll API的功能

相关技能

pytorch

S
toolCode Lib / 代码库
92/ 100

“它是深度学习的瑞士军刀,但祝你好运能从47种安装方法里找到那个不会搞崩你系统的那一个。”

agno

S
toolCode Lib / 代码库
90/ 100

“它承诺成为智能体领域的Kubernetes,但得看开发者有没有耐心学习又一个编排层。”

nuxt-skills

S
toolCo-Pilot / 辅助式
90/ 100

“这本质上是一份组织良好的小抄,能把你的 AI 助手变成一只 Nuxt 框架的复读机。”

免责声明:本内容来源于 GitHub 开源项目,仅供展示和评分分析使用。

版权归原作者所有 janwilmake.