- name
- soul-dreaming
- description
- >
Dreaming — Progressive Memory Management for AI Agents
Prevent amnesia from context compaction. Persist what matters, forget what doesn't.
Memory Architecture (v2 — Categorized Split)
workspace/
└── memory/
├── INDEX.md # 索引 — 唯一必须全量加载的文件
├── decisions.md # 架构决策 + 理由（为什么选 X 不选 Y）
├── lessons.md # 教训与经验（bug 根因、workaround、踩坑记录）
├── project-context.md # 项目上下文（技术栈、目录结构、部署信息）
├── people.md # 人物信息（偏好、角色、沟通习惯）
├── tools-config.md # 工具与环境（命令备忘、环境配置、代理设置）
├── YYYY-MM-DD.md # 每日日记（自动创建）
└── archive/ # 归档（低价值淘汰 + 超期条目）Line Budget per File（每文件行数预算）
| File | Budget | Rationale |
|---|---|---|
| INDEX.md | 80 | Index only — fast to scan |
| decisions.md | 200 | Architecture decisions with rationale |
| lessons.md | 300 | Bug patterns, workarounds — high volume |
| project-context.md | 150 | Stack, schema, config, deploy info |
| people.md | 100 | User preferences, roles |
| tools-config.md | 100 | Commands, env, proxy notes |
| Each daily journal | 100 | Raw events, temporary |
Record Format（标准记录格式）
Every entry in categorized files uses this format:
## [YYYY-MM-DD] Title
- Tags: tag1, tag2, tag3
- Fitness: last_referenced=YYYY-MM-DD, hit_count=N, decision_value=HIGH|MEDIUM|LOW
- Body: one-line summary of the recordThree Stages (Core — 三阶段提炼)
Stage 1: Light Sleep — Every Session End
Scan the current session. Write to today's journal (memory/YYYY-MM-DD.md):
- Decisions made — anything that affects future work
- Key discoveries — bugs found, root causes, unexpected behaviors
- Pending tasks — unfinished items that must survive session restart
- Context anchors — file paths, API endpoints, config values
Iron rule (铁律): Write decisions the moment they're made, not at session end. Context can vanish at any time.
Stage 2: Deep Sleep — Every 3 Days
Promote from daily journals to categorized memory files:
- Scan last 3 days of journals
- Evaluate each item against promotion criteria
- Route to the correct file (decisions → decisions.md, bugs → lessons.md, etc.)
- Delete absorbed journal entries
Promotion criteria:
- Referenced across 2+ sessions/days
- Has ongoing decision value
- Contains a hard-won lesson
- Addresses a recurring pattern
Routing map:
| Journal item type | Target file |
|---|---|
| Architecture/design choice | decisions.md |
| Bug root cause, workaround | lessons.md |
| Tech stack, deploy info | project-context.md |
| User preference, team role | people.md |
| Command, env config | tools-config.md |
Stage 3: REM — Weekly
Memory hygiene via Survival Merge Protocol (优胜劣汰):
- Evaluate fitness of all entries across all files
- Archive low-value entries →
memory/archive/ - Promote high-value — entries hit 5+ times → mark
[PROVEN] - Handle conflicts — supersede old with new, keep one-line summary
- Rebuild INDEX.md — update all stats
Detailed protocol: references/survival-merge.mdMount Strategy（挂载策略 — 三级加载）
Level 1 — Always Mount（常驻加载）
memory/INDEX.md— the only file you must load fully on every wake- Today's
memory/YYYY-MM-DD.md - Yesterday's
memory/YYYY-MM-DD.md
Level 2 — On-Demand Mount（按需加载）
Based on INDEX.md search hints, load when the task touches a domain:
- Starting to write code →
decisions.md+project-context.md - Debugging a bug →
lessons.md - User-facing task →
people.md - Env/setup work →
tools-config.md
Load-and-release: read the file, use it, do NOT carry it into subsequent turns.
Level 3 — On-Demand Search（按需检索）
- Use
memory_search(qmd or equivalent) forarchive/files - For cross-day event lookup without loading full journals
- Never load archive files entirely unless specifically needed
Detailed protocol: references/mount-strategy.mdINDEX.md — Memory Index（索引文件）
INDEX.md is the single source of truth for what memory exists. Rebuild it after every Deep Sleep or REM cycle.
# Memory Index
> Last rebuilt: YYYY-MM-DD | Total files: N | Budget: 500 lines per category
## Hot Files (load on session start)
| File | Lines | Last Updated | Recent Entries |
|------|-------|-------------|----------------|
| decisions.md | 45 | 2026-04-06 | Redis缓存策略, API版本v2 |
| project-context.md | 80 | 2026-04-07 | QMS技术栈, 数据库schema |
## On-Demand Files (load when relevant)
| File | Lines | Last Updated | Search Hint |
|------|-------|-------------|-------------|
| lessons.md | 120 | 2026-04-05 | bug根因, workaround, 踩坑 |
| people.md | 30 | 2026-03-28 | 人物偏好, 角色 |
| tools-config.md | 25 | 2026-04-03 | 命令备忘, 环境配置 |
## Daily Journals
| Date | Exists | Key Events |
|------|--------|------------|
| 2026-04-07 | ✅ | cron修复, Dreaming skill创建 |
| 2026-04-06 | ✅ | PraisonAI研究, GitHub同步 |Session Lifecycle Integration
On Wake (新会话启动)
1. Read memory/INDEX.md
2. Read today + yesterday journals
3. Resume workBefore Context Loss (上下文即将丢失)
→ Dump to today's journal immediately: task state, in-flight decisions, open files, errors, next step.
Quick Reference
| When | Action | Write to |
|---|---|---|
| Decision made | Write immediately | memory/YYYY-MM-DD.md |
| Session ending | Dump current state | memory/YYYY-MM-DD.md |
| New session | Read INDEX + 2 journals | L1 mount |
| Every 3 days | Promote + route + rebuild INDEX | Deep Sleep |
| Weekly | Fitness eval + archive + supersede | REM |
| Task touches domain | Load relevant file | L2 mount |
References
references/memory-lifecycle.md— 三阶段操作手册 + 文件分类规范references/anti-amnesia-checklist.md— 防失忆检查清单 + INDEX 维护references/mount-strategy.md— 挂载策略详细操作手册references/survival-merge.md— 优胜劣汰融合完整协议