feat: OpenClaw integration support with per-skill memory isolation

- Add comprehensive OpenClaw documentation and workflow guides
- Add Linux one-line install script (curl | bash)
- Add example GitHub PR reviewer skill
- Add llms.txt for AI agent documentation
- Update README with OpenClaw quick start (English & Chinese)
- Add per-skill memory isolation using context parameter

This release focuses on making aimemo the go-to memory solution for
OpenClaw skills, with zero-dependency installation and complete isolation.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Author: Panda

CHANGELOG.md

@@ -0,0 +1,100 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [0.4.0] - 2026-02-20
+
+### Added
+
+- **OpenClaw Integration Documentation**
+  - [OpenClaw Integration Guide](docs/openclaw-integration.md) with 5-minute setup
+  - [OpenClaw Workflow](docs/openclaw-workflow.md) with complete architecture diagrams
+  - Per-skill memory isolation using `context` parameter
+  - Comparison with other memory solutions (Cognee, memsearch, Supermemory)
+
+- **Linux Support**
+  - One-line install script (`install.sh`) for Linux amd64/arm64
+  - Automatic architecture detection
+  - Post-install instructions and troubleshooting
+  - Native static binaries (CGO_ENABLED=0)
+
+- **Example OpenClaw Skill**
+  - Complete GitHub PR reviewer skill ([examples/openclaw-github-pr-reviewer/](examples/openclaw-github-pr-reviewer/))
+  - Demonstrates code style preference learning
+  - Shows pattern recognition across sessions
+  - Includes debugging and customization guide
+
+- **AI Agent Documentation**
+  - `llms.txt` file following [llmstxt.org](https://llmstxt.org/) specification
+  - Machine-readable project documentation
+  - Optimized for AI agent consumption
+  - Includes all tools, commands, and integration guides
+
+### Improved
+
+- **README Updates (English & Chinese)**
+  - Added OpenClaw Quick Start section
+  - Added Linux one-line installation option
+  - Added per-skill memory isolation explanation
+  - Added file system layout diagrams
+  - Added comparison table with alternative solutions
+
+- **Documentation Structure**
+  - Created `docs/` directory for comprehensive guides
+  - Created `examples/` directory for working skill implementations
+  - Improved navigation with clear section hierarchy
+
+### Changed
+
+- Quick Start section now shows both Homebrew and curl installation methods
+- Client Support section reorganized for better clarity
+
+### Technical
+
+- Verified Linux amd64/arm64 build compatibility
+- Tested context parameter creates isolated databases correctly
+- Confirmed pure Go SQLite (modernc.org/sqlite) works on Linux
+- Validated goreleaser configuration for multi-platform releases
+
+## [0.3.0] - 2026-02-17
+
+### Added
+
+- Initial public release
+- MCP stdio server implementation
+- Five core memory tools: `memory_context`, `memory_store`, `memory_search`, `memory_forget`, `memory_link`
+- SQLite + FTS5 full-text search with BM25 ranking
+- Importance scoring (recency + access frequency)
+- Context parameter for memory isolation
+- CLI tools for human inspection and management
+- Configuration system (global and per-project)
+- WAL mode for concurrent access
+- Journal entry support
+- Entity and observation management
+- Tag-based organization
+- Relationship linking
+- Export/import (Markdown and JSON)
+- Client support for Claude Code, Cursor, Windsurf, Cline, Continue, Zed
+- Homebrew formula for macOS installation
+- Comprehensive README documentation (English and Chinese)
+- MIT License
+
+### Technical
+
+- Go 1.22+ required
+- Zero runtime dependencies
+- Single binary distribution
+- Pure Go SQLite via modernc.org/sqlite
+- MCP protocol version 2024-11-05
+- 5-second timeout per MCP call
+- < 2000 token schema overhead
+
+---
+
+## Release Links
+
+- [v0.4.0](https://github.com/MyAgentHubs/aimemo/releases/tag/v0.4.0) - OpenClaw Integration Release
+- [v0.3.0](https://github.com/MyAgentHubs/aimemo/releases/tag/v0.3.0) - Initial Release

README.md

@@ -45,6 +45,10 @@ $ claude "let's keep working on the payment service"
 
 ```bash
 # 1. Install
+# Linux/macOS (one-line install):
+curl -sSL https://raw.githubusercontent.com/MyAgentHubs/aimemo/main/install.sh | bash
+
+# Or macOS via Homebrew:
 brew install MyAgentHubs/tap/aimemo
 
 # 2. Initialize memory for your project (run from project root)
@@ -56,6 +60,10 @@ claude mcp add-json aimemo-memory '{"command":"aimemo","args":["serve"]}'
 
 Restart Claude Code. On the next session, Claude will automatically load project context.
 
+### Quick Start for OpenClaw
+
+If you're using OpenClaw skills, see the [OpenClaw Integration](#-openclaw-integration) section below for per-skill memory isolation.
+
 ## 🔧 How It Works
 
 `aimemo serve` runs as a stdio MCP server; Claude Code manages the process lifecycle, so there is nothing to keep alive yourself. When Claude starts a session it calls `memory_context` to load relevant prior context; as it works it calls `memory_store` and `memory_link` to record decisions and relationships. You can call `aimemo search`, `aimemo list`, or `aimemo get` at any time to read or edit the same data from your terminal. Everything lives in a SQLite database inside `.aimemo/`, discovered by walking up from the current directory — the same way Git finds `.git/`.
@@ -157,6 +165,138 @@ This project uses aimemo for persistent memory across sessions.
 - Do not store secrets, credentials, or PII.
 ```
 
+## 🦞 OpenClaw Integration
+
+aimemo solves OpenClaw's "remembers everything but understands none" problem with **per-skill memory isolation** and **zero infrastructure**.
+
+### Why aimemo for OpenClaw?
+
+**The Problem:**
+- OpenClaw's native Markdown memory gets worse the more you use it
+- Skills share memory, causing cross-contamination
+- Context compression loses important context
+
+**The Solution:**
+- ✅ **Zero dependencies** — Single Go binary, no Docker/Node.js/databases
+- ✅ **Per-skill isolation** — Each skill gets its own memory database
+- ✅ **Actually works** — BM25 search + importance scoring finds what matters
+- ✅ **Local-first** — All data stays on your machine
+
+**vs Other Solutions:**
+
+| | aimemo | Cognee | memsearch | Supermemory |
+|---|--------|---------|-----------|-------------|
+| **Dependencies** | Zero | Neo4j/Kuzu | Milvus | Cloud service |
+| **Installation** | 30 sec | Complex | Complex | Sign up required |
+| **Skill isolation** | Built-in | Manual | Manual | N/A |
+| **Linux support** | ✅ Native | ✅ | ✅ | N/A |
+
+### 5-Minute Setup
+
+```bash
+# 1. Install (Linux amd64/arm64)
+curl -sSL https://raw.githubusercontent.com/MyAgentHubs/aimemo/main/install.sh | bash
+
+# 2. Register MCP server with OpenClaw
+claude mcp add-json aimemo-memory '{"command":"aimemo","args":["serve"]}'
+
+# Or add to ~/.openclaw/openclaw.json:
+# {
+#   "mcpServers": {
+#     "aimemo-memory": {
+#       "command": "/usr/local/bin/aimemo",
+#       "args": ["serve"]
+#     }
+#   }
+# }
+
+# 3. Initialize workspace memory
+cd ~/.openclaw/workspace
+aimemo init
+
+# 4. Restart OpenClaw Gateway
+# Linux: systemctl --user restart openclaw-gateway
+# macOS: launchctl stop com.openclaw.gateway && launchctl start com.openclaw.gateway
+```
+
+### Per-Skill Memory Isolation
+
+Each skill gets its own isolated memory by using the `context` parameter:
+
+**In your SKILL.md:**
+```markdown
+---
+name: my-skill
+description: A skill with persistent memory
+---
+
+# My Skill
+
+## Instructions
+
+When doing work:
+
+1. **Load memory FIRST**:
+   ```
+   memory_context({context: "my-skill"})
+   ```
+
+2. Do your task with loaded context
+
+3. **Store learnings**:
+   ```
+   memory_store({
+     context: "my-skill",
+     entities: [{
+       name: "preferences",
+       entityType: "config",
+       observations: ["User prefers snake_case"]
+     }]
+   })
+   ```
+
+**CRITICAL**: Always pass `context: "my-skill"` to prevent memory pollution.
+```
+
+**Result:**
+```
+~/.openclaw/workspace/.aimemo/
+├── memory.db                    # Shared/default (no context)
+├── memory-skill-a.db            # Skill A's isolated memory
+├── memory-skill-b.db            # Skill B's isolated memory
+└── memory-skill-c.db            # Skill C's isolated memory
+```
+
+### Complete Example
+
+See [`examples/openclaw-github-pr-reviewer/`](examples/openclaw-github-pr-reviewer/) for a full working skill that:
+- Reviews GitHub PRs
+- Learns code style preferences
+- Remembers patterns across sessions
+- Stores feedback for improvement
+
+### Documentation
+
+- **[OpenClaw Integration Guide](docs/openclaw-integration.md)** — Step-by-step setup
+- **[OpenClaw Workflow](docs/openclaw-workflow.md)** — Architecture deep-dive
+- **[Example Skill](examples/openclaw-github-pr-reviewer/)** — Complete implementation
+
+### Debugging
+
+```bash
+# List a skill's memory
+aimemo list --context my-skill
+
+# Search within a skill
+aimemo search "keyword" --context my-skill
+
+# Export for inspection
+aimemo export --context my-skill --format json > memory.json
+
+# Get database stats
+aimemo stats --context my-skill
+```
+
 ## 🖥 Client Support
 
 aimemo works with any MCP-compatible AI coding client. The server command is always `aimemo serve`.