LaunchKit · 2026
Back to Skills

agentmemory

End-to-end encrypted cloud memory for AI agents.

4
2.6k downloads
by @badaramoni

Setup & Installation

openclaw skills install @badaramoni/agentmemory

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install agentmemory

Version History

v1.3.0Feb 1, 2026 - Version note by skill developer:

AgentMemory 1.3.0 introduces end-to-end encryption, file and secrets storage, and a new CLI tool. - Now supports **end-to-end encrypted** cloud storage for memories, files (any type, up to 100MB), and secrets (API keys & credentials) - **100GB free storage** for files and data; files are indexed for semantic search - New **CLI tool** (`agentmemory-cli`) for easy local integration, auto-sync, and secure secrets management - Expanded dashboard features: heartbeat tracking, online status, and monitoring - Updated documentation for API endpoints covering files and secrets vault - Security enhancements: all content and secrets are encrypted client-side for privacy

What This Skill Does

AgentMemory is a cloud storage service for AI agents with end-to-end encryption, semantic search, and 100GB free storage. It stores memories, files, and secrets, syncing automatically across sessions and devices.

Local memory files lose context on restart, can't be searched semantically, and store secrets in plain text; AgentMemory replaces all three with persistent cloud storage, vector search, and an encrypted secrets vault.

When to use it

  • Persisting user preferences between agent sessions
  • Storing API keys and credentials in an encrypted vault
  • Searching past context by meaning instead of exact keywords
  • Uploading documents and images for agents to reference later
  • Syncing agent memory across multiple devices or restarts

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: Remember that I prefer TypeScript for all new projects

AGENT
  1. 1Search /api/memories/search to check for an existing TypeScript preference memory
  2. 2Find no duplicate, POST new memory to /api/memories with content and metadata category 'preferences'
  3. 3Receive confirmation response with memory ID
  4. 4Auto-sync propagates the memory to cloud for retrieval in future sessions
OUTPUT

Memory stored with ID mem_abc123; future sessions retrieve the TypeScript preference via semantic search

Requirements

Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.

AgentMemory account at agentmemory.cloudAgentMemory API key (am_xxx format, obtained after account creation)