AgenticMemory

@Tyga.Cloud Ltd

About AgenticMemory

Persistent memory for AI agents - conversation history, key-value context, and semantic search across sessions, plus FIFO queues. 17 MCP tools, remote SSE (nothing to install) or local stdio. Agents self-signup via CLI a

#memory #agent #persistent #semantic-search #context #conversation-history #mcp #ai-agents #cli #redis

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "agenticmemory": {
      "command": "npx",
      "args": [
        "-y",
        "agmry",
        "mcp-serve"
      ],
      "env": {
        "AGMRY_API_KEY": "amk_your_key_here"
      }
    }
  }
}

Tools

Store a message in conversation history. Use for remembering what just happened.

Store multiple messages at once.

Recall recent messages from conversation history. Returns newest first.

Clear all messages in a memory space.

Semantic search across past conversations. Finds by meaning, not keywords.

Set a key-value context entry. Stores structured data (JSON or string).

Get a context value by key.

List all context keys and values in a space.

Delete a context key.

Add a long-term memory entry (summary, decision, extraction, artifact).

List long-term memory entries. Filter by tags.

List all memory spaces owned by this API key.

Create a new memory space.

Enqueue a JSON envelope onto a FIFO queue in a memory space. Use to send work/events to other agents sharing the space.

Dequeue the oldest envelope from a FIFO queue (returns null if empty). Set wait (max 25s) to long-poll for the next envelope.

Peek at a queue without consuming: returns length and the head envelope.

Load full context in one call: recent messages, entries, context, scratchpad. Run this at session start.

Overview

What is Agenticmemory?

Persistent memory for AI agents — conversation history, context, semantic search, queues. CLI + MCP. As easy as git.

How to use Agenticmemory?

Follow the repository README to install the server and add its MCP configuration to your client.

Key features of Agenticmemory

Semantic search is impossible on zero-knowledge spaces by design
Losing the key or passphrase means the data is unrecoverable. That's the point. Back it up: agmry key show --reveal
Conversation history — ordered, role-aware messages with recency windowing, sub-ms reads
Key-value context — typed durable facts: decisions, preferences, runbooks
Long-term entries — titled, tagged knowledge that survives months, with auto-summarisation