Skip to content

Releases: stephenlzc/aiseact

v1.2.0

14 Mar 02:22
@stephenlzc stephenlzc
v1.2.0
cff6953

Choose a tag to compare

View all tags
v1.2.0 Latest
Latest

What's New

✨ Features

  • Integrated multi-search-engine skill with 17 search engines (8 CN + 9 Global)
  • Added Search Tools section to SKILL.md with engine selection guide
  • Created comprehensive multi-engine guide and unified search syntax reference

📚 Documentation

  • New - Engine selection matrix and usage patterns
  • New - Unified search syntax combining AISEACT + multi-search-engine
  • Updated workflow.md and search-strategies.md with multi-engine strategies
  • Added multi-engine research case study to case-studies.md

🔧 Configuration

  • Added multi-search-engine integration options to CONFIGURATION.md

🙏 Acknowledgments

  • Added thanks to @gpyAngyoujun for the brilliant idea of aggregating 17 search engines

In the vibe coding era, IDEAS ARE THE REAL ENGINE

Assets 2
Loading

AISEACT v1.1.0

13 Mar 13:54
@stephenlzc stephenlzc
1.1.0
93455b5

Choose a tag to compare

View all tags
AISEACT v1.1.0

📊 AISEACT v1.1.0 - Performance Evaluation & Optimization Update

English Release Notes

✨ New Features

  • EVALUATION_REPORT.md: Comprehensive performance evaluation report with quantitative metrics
    • P0 source usage increased from 5-15% to 85-100%
    • 75% time savings on source verification
    • 83% reduction in fact-check error rate
  • CONFIGURATION.md: Detailed configuration guide for customization
  • SECURITY.md: Security documentation
  • TRUST.md: Trust and transparency report
  • references/quick-reference.md: Bilingual quick reference for source ratings

🔧 Improvements

  • SKILL.md Optimization: Reduced from 311 lines to 72 lines for faster loading
  • README Enhancement: Added benefits table showing key performance metrics
  • Source References: Updated unreliable-sources.md with Wikipedia reliable sources references

📖 Documentation

  • Added detailed evaluation report with test scenarios
  • Bilingual quick-reference guide (Chinese sources in Chinese, English sources in English)
  • Comprehensive trust and transparency documentation

中文发布说明

✨ 新增功能

  • EVALUATION_REPORT.md: 包含量化指标的综合性能评估报告
    • P0 来源使用率从 5-15% 提升至 85-100%
    • 来源验证时间节省 75%
    • 事实核查错误率降低 83%
  • CONFIGURATION.md: 详细的配置指南
  • SECURITY.md: 安全文档
  • TRUST.md: 信任与透明度报告
  • references/quick-reference.md: 来源评级的双语快速参考

🔧 改进

  • SKILL.md 优化: 从 311 行精简至 72 行，加载更快
  • README 增强: 添加了展示关键性能指标的优势表格
  • 来源参考: 更新了 unreliable-sources.md，添加了维基百科可靠来源参考

📖 文档

  • 添加了包含测试场景的详细评估报告
  • 双语快速参考指南（中文来源用中文，英文来源用英文）
  • 全面的信任与透明度文档

🔗 Key Links

✅ Upgrade Recommended

This release includes significant performance improvements and comprehensive documentation. All users are recommended to upgrade.

推荐升级: 此版本包含显著的性能改进和全面的文档。建议所有用户升级。

Loading

AISEACT v1.0.0 - Initial Release

13 Mar 10:43
@stephenlzc stephenlzc
v1.0.0
08d2afa

Choose a tag to compare

View all tags
AISEACT v1.0.0 - Initial Release

🎉 AISEACT v1.0.0 - Initial Release

AI Search Enhancement and Cross-verification Tool

What is AISEACT?

AISEACT (AI-Search-EhAnCemenT) is a search enhancement skill for AI agents (OpenClaw, KimiClaw, MaxClaw, Claude Code, etc.) that solves the problem of "high search capability but low result quality".

🌟 Key Features

  • Primary Source First: Prioritizes official documents, government announcements, and authoritative sources
  • 6-Phase Iterative Search: Think → Search Round 1 → Analyze Gaps → Search Round 2 → Cross-Validate → Answer
  • Content Farm Blacklist: 1000+ unreliable sources including Baijiahao, Sohu, CSDN scrapers
  • Cross-Validation Framework: Ensures information accuracy through multiple independent sources
  • Multi-Language Support: Full documentation in English and Chinese

📁 Project Structure

aiseact/
├── README.md                    # English documentation
├── README.zh-CN.md              # Chinese documentation
├── SKILL.md                     # Main skill file for AI agents
└── references/
    ├── unreliable-sources.md    # Content farm blacklist
    ├── authority-sources.md     # Authoritative source recommendations
    ├── workflow.md              # Detailed 6-phase workflow
    ├── search-strategies.md     # Search syntax & techniques
    └── case-studies.md          # Real-world examples

🚀 Quick Start

Copy and paste this prompt to your LLM agent:

Install the AISEACT skill to help me with search tasks.
Clone from: https://github.com/stephenlzc/aiseact
Use the SKILL.md to guide all web searches - prioritize primary sources, avoid content farms, and cross-validate important information.

📝 Release Notes

This is the initial release of AISEACT, providing a complete methodology for improving AI search quality.

Full Changelog: https://github.com/stephenlzc/aiseact/commits/v1.0.0

Loading