🌐 简体中文 | 繁體中文 | English | Español | Deutsch | Français | 日本語
Give your AI coding assistant a memory — Cross-session persistent memory MCP Server
Still using CLAUDE.md / MEMORY.md as memory? This Markdown-file memory approach has fatal flaws: the file keeps growing, injecting everything into every session and burning massive tokens; content only supports keyword matching — search "database timeout" and you won't find "MySQL connection pool pitfall"; sharing one file across projects causes cross-contamination; there's no task tracking, so dev progress lives entirely in your head; not to mention the 200-line truncation, manual maintenance, and inability to deduplicate or merge.
AIVectorMemory is a fundamentally different approach. Local vector database storage with semantic search for precise recall (matches even when wording differs), on-demand retrieval that loads only relevant memories (token usage drops 50%+), automatic multi-project isolation with zero interference, and built-in issue tracking + task management that lets AI fully automate your dev workflow. All data is permanently stored on your machine — zero cloud dependency, never lost when switching sessions or IDEs.
| Feature | Description |
|---|---|
| 🧠 Cross-Session Memory | Your AI finally remembers your project — pitfalls, decisions, conventions all persist across sessions |
| 🔍 Semantic Search | No need to recall exact wording — search "database timeout" and find "MySQL connection pool issue" |
| 💰 Save 50%+ Tokens | Stop copy-pasting project context every conversation. Semantic retrieval on demand, no more bulk injection |
| 🔗 Task-Driven Dev | Issue tracking → task breakdown → status sync → linked archival. AI manages the full dev workflow |
| 📊 Desktop App + Web Dashboard | Native desktop app (macOS/Windows/Linux) + Web dashboard, visual management for memories and tasks, 3D vector network reveals knowledge connections at a glance |
| 🏠 Fully Local | Zero cloud dependency. ONNX local inference, no API Key, data never leaves your machine |
| 🔌 All IDEs | Cursor / Kiro / Claude Code / Windsurf / VSCode / OpenCode / Trae — one-click install, works out of the box |
| 📁 Multi-Project Isolation | One DB for all projects, auto-isolated with zero interference, seamless project switching |
| 🔄 Smart Dedup | Similarity > 0.95 auto-merges updates, keeping your memory store clean — never gets messy over time |
| 🌐 7 Languages | 简体中文 / 繁體中文 / English / Español / Deutsch / Français / 日本語, full-stack i18n for dashboard + Steering rules |
QQ群：1085682431 | 微信：changhuibiz
共同参与项目开发加QQ群或微信交流
Login
Project Selection
Overview & Vector Network
┌─────────────────────────────────────────────────┐
│ AI IDE │
│ OpenCode / Claude Code / Cursor / Kiro / ... │
└──────────────────────┬──────────────────────────┘
│ MCP Protocol (stdio)
┌──────────────────────▼──────────────────────────┐
│ AIVectorMemory Server │
│ │
│ ┌──────────┐ ┌──────────┐ ┌──────────────────┐ │
│ │ remember │ │ recall │ │ auto_save │ │
│ │ forget │ │ task │ │ status/track │ │
│ └────┬─────┘ └────┬─────┘ └───────┬──────────┘ │
│ │ │ │ │
│ ┌────▼────────────▼───────────────▼──────────┐ │
│ │ Embedding Engine (ONNX) │ │
│ │ intfloat/multilingual-e5-small │ │
│ └────────────────────┬───────────────────────┘ │
│ │ │
│ ┌────────────────────▼───────────────────────┐ │
│ │ SQLite + sqlite-vec (Vector Index) │ │
│ │ ~/.aivectormemory/memory.db │ │
│ └────────────────────────────────────────────┘ │
└──────────────────────────────────────────────────┘
# Install
pip install aivectormemory
# Upgrade to latest version
pip install --upgrade aivectormemory
# Navigate to your project directory, one-click IDE setup
cd /path/to/your/project
run installrun install interactively guides you to select your IDE, auto-generating MCP config, Steering rules, and Hooks — no manual setup needed.
macOS users note:
- If you get
externally-managed-environmenterror, add--break-system-packages- If you get
enable_load_extensionerror, your Python doesn't support SQLite extension loading (macOS built-in Python and python.org installers don't support it). Use Homebrew Python instead:brew install python /opt/homebrew/bin/python3 -m pip install aivectormemory
No pip install needed, run directly:
cd /path/to/your/project
uvx aivectormemory installRequires uv to be installed.
uvxauto-downloads and runs the package — no manual installation needed.
{
"mcpServers": {
"aivectormemory": {
"command": "run",
"args": ["--project-dir", "/path/to/your/project"]
}
}
}📍 IDE Configuration File Locations
| IDE | Config Path |
|---|---|
| Kiro | .kiro/settings/mcp.json |
| Cursor | .cursor/mcp.json |
| Claude Code | .mcp.json |
| Windsurf | .windsurf/mcp.json |
| VSCode | .vscode/mcp.json |
| Trae | .trae/mcp.json |
| OpenCode | opencode.json |
content (string, required) Memory content in Markdown format
tags (string[], required) Tags, e.g. ["pitfall", "python"]
scope (string) "project" (default) / "user" (cross-project)
Similarity > 0.95 auto-updates existing memory, no duplicates.
query (string) Semantic search keywords
tags (string[]) Exact tag filter
scope (string) "project" / "user" / "all"
top_k (integer) Number of results, default 5
Vector similarity matching — finds related memories even with different wording.
memory_id (string) Single ID
memory_ids (string[]) Batch IDs
state (object, optional) Omit to read, pass to update
is_blocked, block_reason, current_task,
next_step, progress[], recent_changes[], pending[]
Maintains work progress across sessions, auto-restores context in new sessions.
action (string) "create" / "update" / "archive" / "list"
title (string) Issue title
issue_id (integer) Issue ID
status (string) "pending" / "in_progress" / "completed"
content (string) Investigation content
action (string, required) "batch_create" / "update" / "list" / "delete" / "archive"
feature_id (string) Linked feature identifier (required for list)
tasks (array) Task list (batch_create, supports subtasks)
task_id (integer) Task ID (update)
status (string) "pending" / "in_progress" / "completed" / "skipped"
Links to spec docs via feature_id. Update auto-syncs tasks.md checkboxes and linked issue status.
action (string) "generate" (default) / "diff" (compare differences)
lang (string) Language: en / zh-TW / ja / de / fr / es
sections (string[]) Specify sections: header / tools / deps
Auto-generates README content from TOOL_DEFINITIONS / pyproject.toml, multi-language support.
preferences (string[]) User-expressed technical preferences (fixed scope=user, cross-project)
extra_tags (string[]) Additional tags
Auto-extracts and stores user preferences at end of each conversation, smart dedup.
run web --port 9080
run web --port 9080 --quiet # Suppress request logs
run web --port 9080 --quiet --daemon # Run in background (macOS/Linux)Visit http://localhost:9080 in your browser. Default username admin, password admin123 (can be changed in settings after first login).
- Multi-project switching, memory browse/search/edit/delete/export/import
- Semantic search (vector similarity matching)
- One-click project data deletion
- Session status, issue tracking
- Tag management (rename, merge, batch delete)
- Token authentication protection
- 3D vector memory network visualization
- 🌐 Multi-language support (简体中文 / 繁體中文 / English / Español / Deutsch / Français / 日本語)
Scan to join WeChat group | Scan to join QQ group
AIVectorMemory is the storage layer. Use Steering rules to tell AI when and how to call these tools.
Running run install auto-generates Steering rules and Hooks config — no manual setup needed.
| IDE | Steering Location | Hooks |
|---|---|---|
| Kiro | .kiro/steering/aivectormemory.md |
.kiro/hooks/*.hook |
| Cursor | .cursor/rules/aivectormemory.md |
.cursor/hooks.json |
| Claude Code | CLAUDE.md (appended) |
.claude/settings.json |
| Windsurf | .windsurf/rules/aivectormemory.md |
.windsurf/hooks.json |
| VSCode | .github/copilot-instructions.md (appended) |
.claude/settings.json |
| Trae | .trae/rules/aivectormemory.md |
— |
| OpenCode | AGENTS.md (appended) |
.opencode/plugins/*.js |
📋 Steering Rules Example (auto-generated)
# AIVectorMemory - Workflow Rules
## 1. New Session Startup (execute in order)
1. `recall` (tags: ["project-knowledge"], scope: "project", top_k: 100) load project knowledge
2. `recall` (tags: ["preference"], scope: "user", top_k: 20) load user preferences
3. `status` (no state param) read session state
4. Blocked → report and wait; Not blocked → enter processing flow
## 2. Message Processing Flow
- Step A: `status` read state, wait if blocked
- Step B: Classify message type (chat/correction/preference/code issue)
- Step C: `track create` record issue
- Step D: Investigate (`recall` pitfalls + read code + find root cause)
- Step E: Present plan to user, set blocked awaiting confirmation
- Step F: Modify code (`recall` pitfalls before changes)
- Step G: Run tests to verify
- Step H: Set blocked awaiting user verification
- Step I: User confirms → `track archive` + clear block
## 3. Blocking Rules
Must `status({ is_blocked: true })` when proposing plans or awaiting verification.
Only clear after explicit user confirmation. Never self-clear.
## 4-9. Issue Tracking / Code Checks / Spec Task Mgmt / Memory Quality / Tool Reference / Dev Standards
(Full rules auto-generated by `run install`)🔗 Hooks Config Example (Kiro only, auto-generated)
Auto-save on session end removed. Dev workflow check (.kiro/hooks/dev-workflow-check.kiro.hook):
{
"enabled": true,
"name": "Dev Workflow Check",
"version": "1",
"when": { "type": "promptSubmit" },
"then": {
"type": "askAgent",
"prompt": "Core principles: verify before acting, no blind testing, only mark done after tests pass"
}
}The embedding model (~200MB) is auto-downloaded on first run. If slow:
export HF_ENDPOINT=https://hf-mirror.comOr add env to MCP config:
{
"env": { "HF_ENDPOINT": "https://hf-mirror.com" }
}| Component | Technology |
|---|---|
| Runtime | Python >= 3.10 |
| Vector DB | SQLite + sqlite-vec |
| Embedding | ONNX Runtime + intfloat/multilingual-e5-small |
| Tokenizer | HuggingFace Tokenizers |
| Protocol | Model Context Protocol (MCP) |
| Web | Native HTTPServer + Vanilla JS |
Performance: ONNX INT8 Quantization
- ⚡ Embedding model auto-quantized from FP32 to INT8 on first load, model file from 448MB down to 113MB
- ⚡ MCP Server memory usage reduced from ~1.6GB to ~768MB (50%+ reduction)
- ⚡ Quantization is transparent to users — automatic on first use, cached for subsequent loads, falls back to FP32 on failure
New: Remember Password
- 🔐 Login page on both desktop and web dashboard now has a "Remember password" checkbox
- 🔐 When checked, credentials are saved to localStorage and auto-filled on next login; when unchecked, saved credentials are cleared
- 🔐 Checkbox is hidden in registration mode
Enhancement: Steering Rules
- 📝 IDENTITY & TONE section strengthened with more specific constraints (no pleasantries, no translating user messages, etc.)
- 📝 Self-testing requirements now distinguish between backend-only, MCP Server, and frontend-visible changes (Playwright required for frontend)
- 📝 Development rules now mandate self-testing after completing development
- 📝 All 7 language versions synchronized
- 🐛 Desktop app version comparison switched to semantic versioning, fixing false upgrade prompts when local version is higher
- 🐛 Health check page field names aligned with backend, fixing consistency status always showing Mismatch
- 🔧 check_track.sh hook adds Python fallback, resolving silent hook failure when system sqlite3 is unavailable (#4)
- 🖥️ Desktop app one-click install + upgrade detection
- 🖥️ Auto-detect Python and aivectormemory installation status on startup
- 🖥️ Show one-click install button when not installed, check PyPI and desktop new versions when installed
- 🐛 Installation detection switched to importlib.metadata.version() for accurate package version
- 🔧 Fix PyPI package size anomaly (sdist from 32MB down to 230KB), excluded accidentally packaged dev files
New: Native Desktop App
- 🖥️ Native desktop client supporting macOS (ARM64), Windows (x64), Linux (x64)
- 🖥️ Desktop app shares the same database as Web dashboard, fully feature-equivalent
- 🖥️ Dark/light theme switching, Glass frosted visual style
- 🖥️ Login auth, project selection, stats overview, memory management, issue tracking, task management, tag management, settings, data maintenance — full feature coverage
- 📦 Auto-published installers via GitHub Releases, download and use
New: CI/CD Auto Build
- 🔄 GitHub Actions auto-builds desktop installers for all 3 platforms
- 🔄 Push a tag to trigger the full compile, package, and release pipeline
Fixes
- 🐛 Windows platform compatibility fixes
- 🐛 sqlite-vec extension download URL fix
Optimization: Token Usage Reduction
- ⚡ Steering rules changed from per-message dynamic injection to static loading, reducing repeated token consumption
- ⚡ Greatest impact for Claude Code users — ~2K fewer tokens per message
New: Full-Stack i18n (7 Languages)
- 🌐 Web dashboard + desktop UI fully supports 7 languages: 简体中文 / 繁體中文 / English / Español / Deutsch / Français / 日本語
- 🌐 One-click language switch in settings page, takes effect immediately
- 🌐 MCP tool responses follow language setting, AI replies automatically use the corresponding language
- 🌐 Switching language auto-regenerates steering rules for all installed projects
New: Web Dashboard Settings Page
- ⚙️ Language switch, theme settings, system info display
- ⚙️ Database health check, repair, backup and other maintenance tools
Optimization: Memory Search
- 🔍
recallsearch supports OR/AND tag matching modes, fixing missed results with multi-tag searches - 🔍 Semantic search + tag filter defaults to OR matching (broader), tags-only browsing keeps AND matching (more precise)
📋 v0.2.x and earlier changelog
Apache-2.0