skill/SKILL.md
# AVM Memory Skill AI Virtual Memory for agents. Token-aware recall, knowledge graphs, multi-agent support. ## Quick Start ### 1. Configure AVM Daemon ```yaml # ~/.config/avm/mounts.yaml mounts: - mountpoint: ~/.openclaw/workspace-{agent}/avm agent_id: {agent} ``` ```bash # Start daemon avm-daemon start --daemon # Verify ls ~/.openclaw/workspace-{agent}/avm/ ``` ### 2. Basic Usage ```bash # Store a memory echo "RSI > 70 indicates overbought" > avm/notes/rsi_rule.md # Recall with q
npx skillsauth add aivmem/avm skillInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
AI Virtual Memory for agents. Token-aware recall, knowledge graphs, multi-agent support.
# ~/.config/avm/mounts.yaml
mounts:
- mountpoint: ~/.openclaw/workspace-{agent}/avm
agent_id: {agent}
# Start daemon
avm-daemon start --daemon
# Verify
ls ~/.openclaw/workspace-{agent}/avm/
# Store a memory
echo "RSI > 70 indicates overbought" > avm/notes/rsi_rule.md
# Recall with query
cat "avm/:recall?q=RSI trading"
# See what's in memory
cat avm/:topics
# Browse structure
cat avm/:list
| Path | Description |
|------|-------------|
| avm/:recall?q=... | Semantic search |
| avm/:topics | Topic overview |
| avm/:list | List all memories |
| avm/:handlers | Available handlers |
| avm/:handlers/{name} | Specific handler docs |
| avm/notes/*.md | Read/write memories |
When spawning a sub-agent that needs memory:
# ~/.config/avm/mounts.yaml
mounts:
- mountpoint: ~/.openclaw/workspace-{subagent}/avm
agent_id: {subagent}
# Reload daemon
avm-daemon reload
## Memory
This agent has AVM memory mounted at `avm/`.
### Recall
\`\`\`bash
cat "avm/:recall?q=your query"
\`\`\`
### Remember
\`\`\`bash
echo "content" > avm/notes/title.md
\`\`\`
### Topics
\`\`\`bash
cat avm/:topics
\`\`\`
# Test recall
cat "avm/:recall?q=test"
# Daemon management
avm-daemon start --daemon # Start in background
avm-daemon status # Show mounts
avm-daemon reload # Reload config
avm-daemon stop # Stop all
# CLI operations
avm recall "query" --agent {id}
avm remember "content" --title "name" --agent {id}
avm topics --agent {id}
# Import existing memories
avm import /path/to/*.md --agent {id}
Multiple agents can share memories:
# Agent A writes to shared
echo "market signal" > avm/shared/market/signal.md
# Agent B reads shared
cat "avm/:recall?q=market signal" # Finds it
Each agent's writes go to their private namespace by default:
avm/
├── private/{agent}/*.md # Only this agent
└── shared/{namespace}/*.md # All agents
cat avm/:recall?q=... ranks by relevance + importancecat shared/file.md:delta shows only changes since last readtags: [trading, risk]avm link to build knowledge graphcat avm/:topics to see what's available# Check daemon status
avm-daemon status
# Restart daemon
avm-daemon stop && avm-daemon start --daemon
# Force unmount (macOS)
umount -f /path/to/mount
# Force unmount (Linux)
fusermount -uz /path/to/mount
# Check ownership
ls -la ~/.local/share/avm/
# Fix if needed
chown -R $USER ~/.local/share/avm/
# Check if memories exist
avm list --agent {id}
# Check topics
cat avm/:topics
testing
Create, edit, improve, or audit AgentSkills. Use when creating a new skill from scratch or when asked to improve, review, audit, tidy up, or clean up an existing skill or SKILL.md file. Also use when editing or restructuring a skill directory (moving files to references/ or scripts/, removing stale content, validating against the AgentSkills spec). Triggers on phrases like "create a skill", "author a skill", "tidy up a skill", "improve this skill", "review the skill", "clean up the skill", "audit the skill".
testing
Host security hardening and risk-tolerance configuration for OpenClaw deployments. Use when a user asks for security audits, firewall/SSH/update hardening, risk posture, exposure review, OpenClaw cron scheduling for periodic checks, or version status checks on a machine running OpenClaw (laptop, workstation, Pi, VPS).
testing
Create, edit, improve, or audit AgentSkills. Use when creating a new skill from scratch or when asked to improve, review, audit, tidy up, or clean up an existing skill or SKILL.md file. Also use when editing or restructuring a skill directory (moving files to references/ or scripts/, removing stale content, validating against the AgentSkills spec). Triggers on phrases like "create a skill", "author a skill", "tidy up a skill", "improve this skill", "review the skill", "clean up the skill", "audit the skill".
testing
Host security hardening and risk-tolerance configuration for OpenClaw deployments. Use when a user asks for security audits, firewall/SSH/update hardening, risk posture, exposure review, OpenClaw cron scheduling for periodic checks, or version status checks on a machine running OpenClaw (laptop, workstation, Pi, VPS).