memory-system自研三层记忆恢复系统。解决session重启后"忘记一切"的问题,提供永久记忆、今日记忆、临时记忆的完整架构。
Install via ClawdBot CLI:
clawdbot install daoistbro/memory-system解决的核心问题:Session重启后上下文丢失,像"睡醒后忘记一切"
解决方案:三层记忆恢复系统
存储:
identity.md - 身份、偏好、关系technical-stack.md - 技术栈、工具working-directory.md - 工作目录、习惯key-decisions.md - 关键决策、教训特点:
存储:
YYYY-MM-DD.md - 每日工作、决策、重要事件session-N.md - 每个session的记录特点:
存储:
特点:
1. 读取 memory/YYYY-MM-DD.md(今日日志)
2. 读取 MEMORY.md(全局长期记忆)
3. 读取 memory/permanent/*.md(分类记忆)
4. memory_search定位相关记忆
5. 恢复工作状态
bash /data/workspace/scripts/memory-recovery.sh
✅ 应该记录:
❌ 不应该记录:
我的优势:
向量系统的优势:
混合方案:短期使用文件系统,长期可以结合向量搜索
作者:大哥 (Big Brother)
创建时间:2026-02-11
版本:v1.0.0
许可证:MIT
Generated Mar 1, 2026
AI developers and researchers can use this memory system to maintain context across sessions when testing or fine-tuning models, ensuring continuity in experiments and decision logs without relying on cloud services. It helps track technical stack changes and key decisions locally, reducing dependency on external APIs.
Freelancers handling multiple client projects can leverage the system to store permanent preferences, technical details, and daily logs, enabling quick recovery of work states after breaks or system restarts. This improves productivity by minimizing context-switching overhead and maintaining a clear audit trail.
Instructors and students in coding bootcamps or online courses can use it to preserve learning progress, technical notes, and key lessons across sessions, facilitating personalized learning paths and review. The local storage ensures privacy and control over educational data without internet reliance.
Small business owners or teams can adopt this system to document operational decisions, customer preferences, and daily workflows in a structured manner, aiding in knowledge retention and onboarding new members. It supports recovery from data loss incidents through manual backup processes.
Offer the memory system as a free, open-source package under MIT license, while generating revenue through paid consulting, customization services, and premium support for enterprise clients. This model builds community trust and allows monetization from advanced features or integrations.
Develop a cloud-based version that integrates with popular AI platforms and productivity tools, charging subscription fees for enhanced features like automated backups, cross-device sync, and advanced analytics. This targets users seeking scalability beyond local storage limitations.
Create paid workshops, online courses, and certification programs teaching users how to implement and optimize the memory system for specific industries like AI development or project management. Revenue comes from course fees and certification renewals, leveraging the system's practical utility.
💬 Integration Tip
Integrate with existing workflow tools by setting up session hooks for automatic saves and using the memory_search function to enhance context retrieval in AI assistants or note-taking apps.
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Search and analyze your own session logs (older/parent conversations) using jq.
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linking related objects, enforcing constraints, planning multi-step actions as graph transformations, or when skills need to share state. Trigger on "remember", "what do I know about", "link X to Y", "show dependencies", entity CRUD, or cross-skill data access.
Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vibe-coding ready.
Headless browser automation CLI optimized for AI agents with accessibility tree snapshots and ref-based element selection