ys/GenericAgent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs: fuzzy WeChat references in README

Browse Source
This commit is contained in:
Liang Jiaqing
2026-02-07 08:45:59 +08:00
parent e97dc26a3c
commit 4c203ad748
1 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
README.md
View File

@@ -12,12 +12,12 @@ This agent is not limited to a fixed set of features. Its true power lies in its
- **Self-Discovery via Long-Term Memory**: - **Self-Discovery via Long-Term Memory**:
- The agent maintains a "Global Memory" (L2 Facts) to store system paths, credentials, and environmental status. - The agent maintains a "Global Memory" (L2 Facts) to store system paths, credentials, and environmental status.
- It can autonomously retrieve context-aware SOPs (Standard Operating Procedures) to handle specialized tasks like WeChat database decryption or Gmail API operations. - It can autonomously retrieve context-aware SOPs (Standard Operating Procedures) to handle specialized tasks like Instant Messaging (IM) database recovery or Gmail API operations.
- **Dynamic Tool Manufacturing**: - **Dynamic Tool Manufacturing**:
- Through `code_run`, the agent can write and execute arbitrary Python scripts to interface with new hardware or software. - Through `code_run`, the agent can write and execute arbitrary Python scripts to interface with new hardware or software.
- Examples of self-integrated capabilities include: - Examples of self-integrated capabilities include:
- **Deep Web Interaction**: JS injection via Tampermonkey for UI automation. - **Deep Web Interaction**: JS injection via Tampermonkey for UI automation.
- **Digital Forensics**: Querying SQLCipher-encrypted databases (e.g., WeChat v4.0+). - **Digital Forensics**: Querying SQLCipher-encrypted databases (e.g., encrypted local storage of IM apps).
- **Vision-Driven Logic**: Understanding UI states through local vision APIs (`ask_vision`). - **Vision-Driven Logic**: Understanding UI states through local vision APIs (`ask_vision`).
- **System Indexing**: Utilizing **Everything CLI (es.exe)** for instant file discovery across the entire OS. - **System Indexing**: Utilizing **Everything CLI (es.exe)** for instant file discovery across the entire OS.
- **Android Automation**: ADB-based control for mobile device interaction. - **Android Automation**: ADB-based control for mobile device interaction.
@@ -35,7 +35,7 @@ This agent is not limited to a fixed set of features. Its true power lies in its
"Scan my local memory for recent SOPs regarding mail processing, then find and download my latest reimbursement receipts from Gmail." "Scan my local memory for recent SOPs regarding mail processing, then find and download my latest reimbursement receipts from Gmail."
### 2. Complex Multi-Step Automation ### 2. Complex Multi-Step Automation
"Locate the WeChat database, decrypt it to find messages about 'Project X', and summarize the findings into a PDF report." "Locate a specific encrypted IM database, decrypt it to find messages about 'Project X', and summarize the findings into a PDF report."
### 3. Real-Time System Intervention ### 3. Real-Time System Intervention
"Monitor my cloud dashboard via the browser; if the status turns red, execute a local PowerShell script to restart the service and notify me." "Monitor my cloud dashboard via the browser; if the status turns red, execute a local PowerShell script to restart the service and notify me."
@@ -65,12 +65,12 @@ pc-agent-loop 是一个极致简约的 PC 级自主 AI Agent 框架。它通过
- **基于长期记忆的自我发现**: - **基于长期记忆的自我发现**:
- Agent 维护“全局记忆”（L2 Facts）以存储系统路径、凭据和环境状态。 - Agent 维护“全局记忆”（L2 Facts）以存储系统路径、凭据和环境状态。
- 能够自主检索上下文相关的 SOP（标准作业程序），以处理微信数据库解密、Gmail API 操作等专业任务。 - 能够自主检索上下文相关的 SOP（标准作业程序），以处理即时通讯软件（IM）数据库恢复、Gmail API 操作等专业任务。
- **动态工具制造**: - **动态工具制造**:
- 通过 `code_run`，Agent 可以编写并执行 Python/PowerShell 脚本来对接新硬件或软件。 - 通过 `code_run`，Agent 可以编写并执行 Python/PowerShell 脚本来对接新硬件或软件。
- **自集成能力示例**: - **自集成能力示例**:
- **深度 Web 自动化**: 通过 Tampermonkey 进行 JS 注入实现 UI 自动化。 - **深度 Web 自动化**: 通过 Tampermonkey 进行 JS 注入实现 UI 自动化。
- **数字取证**: 查询 SQLCipher 加密的数据库（如微信 v4.0+）。 - **数字取证**: 查询 SQLCipher 加密的数据库（如加密的本地 IM 数据库）。
- **视觉驱动逻辑**: 通过本地视觉 API (`ask_vision`) 理解 UI 状态。 - **视觉驱动逻辑**: 通过本地视觉 API (`ask_vision`) 理解 UI 状态。
- **系统全盘索引**: 利用 **Everything CLI (es.exe)** 实现毫秒级文件检索。 - **系统全盘索引**: 利用 **Everything CLI (es.exe)** 实现毫秒级文件检索。
- **安卓自动化**: 基于 ADB 控制移动设备交互。 - **安卓自动化**: 基于 ADB 控制移动设备交互。
@@ -85,7 +85,7 @@ pc-agent-loop 是一个极致简约的 PC 级自主 AI Agent 框架。它通过
## 🛠️ 典型使用场景 ## 🛠️ 典型使用场景
1. **环境自适应**: “扫描我的本地记忆寻找邮件处理 SOP，然后从 Gmail 下载最新的报销收据。” 1. **环境自适应**: “扫描我的本地记忆寻找邮件处理 SOP，然后从 Gmail 下载最新的报销收据。”
2. **跨模块协作**: “定位微信数据库并解密，查找关于‘项目 X’的消息，并汇总成 PDF 报告。” 2. **跨模块协作**: “定位特定的加密 IM 数据库并解密，查找关于‘项目 X’的消息，并汇总成 PDF 报告。”
3. **系统干预**: “监控云端控制台，若状态异常则执行本地脚本重启服务并邮件通知我。” 3. **系统干预**: “监控云端控制台，若状态异常则执行本地脚本重启服务并邮件通知我。”
## 🧩 7 大核心原子工具 ## 🧩 7 大核心原子工具