Memory Pro (v2)

This skill provides semantic search over your memory files using a local vector database.

Architecture (v2)

• Service: Runs as a systemd user service (memory-pro.service).
• Port: 8001 (hardcoded for stability).
• Engine: FAISS + Sentence-Transformers (all-MiniLM-L6-v2).
• Data Source:
  • Daily logs: ${OPENCLAW_WORKSPACE}/memory/*.md
  • Core files: MEMORY.md, SOUL.md, STATUS.md, AGENTS.md, USER.md (from workspace root).
• Index: Stored in ${OPENCLAW_WORKSPACE}/skills/memory-pro/v2/memory.index.

Usage

1. Semantic Search (Recommended)

Use the python script to query the running service.

# Basic search
python3 scripts/search_semantic.py "What did I do yesterday?"

# JSON output
python3 scripts/search_semantic.py "project updates" --json

2. Manual Index Rebuild

The service automatically rebuilds the index on restart. To force an update:

systemctl --user restart memory-pro.service

Note: Service restart takes ~15-20 seconds to rebuild index and load models. The client script has auto-retry logic.

3. Service Management

# Check status
systemctl --user status memory-pro.service

# Stop service
systemctl --user stop memory-pro.service

# View logs
journalctl --user -u memory-pro.service -f

Troubleshooting

"Connection failed"

• The service might be stopped or restarting.
• Check status: systemctl --user status memory-pro.service.
• If restarting, wait 15 seconds. The client script retries automatically for up to 20s.

"Index size mismatch"

• This means memory.index and sentences.txt are out of sync.
• Fix: Restart the service. The startup script start.sh automatically runs build_index.py to fix this consistency issue before starting the API.

"Address already in use"

• Port 8001 is taken by a zombie process.
• Fix: kill $(lsof -t -i:8001) then restart service.