openviking
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@openvikingsearch my memories for anything about the database connection issue"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
OpenViking Plugin for Google Antigravity CLI (agy)
Official OpenViking long-term memory and context database plugin for Google Antigravity CLI (agy).
This plugin provides 100% feature parity with the Claude Code and Codex OpenViking memory plugins, giving Antigravity seamless, zero-friction long-term memory across sessions.
π Key Features
π§ Passive Auto-Recall (
PreInvocation): Automatically searches OpenViking memories and injects relevant historical facts, conventions, and user preferences into the model prompt before each turn.π Initial Profile Injection: Dynamically injects your OpenViking user profile, memory index, and
<available-skills>catalog on session startup.π‘οΈ Virtual URI Guard (
PreToolUse): Prevents native filesystem tools (view_file,write_to_file, etc.) from corrupting virtualviking://URIs and automatically redirects the agent to OpenViking MCP tools.π₯ Incremental Auto-Capture (
PostInvocation): Stream-captures finished conversation turns without interrupting the model or polluting context.πΎ Automatic Session Commit (
Stop): Automatically submits conversation transcripts to OpenViking on task exit for asynchronous memory and lesson extraction.π οΈ Full MCP Toolset: Connects to the local OpenViking daemon over stdio, exposing 15 core tools (
find,search,read,write,edit,remember,tree,add_resource, etc.).π Built-in Agent Skills: Bundles 4 expert skills:
openviking-memory,openviking-skills,ov-experience-memory, andov-memory-troubleshoot.
Related MCP server: AI Agent History RAG MCP Server
ποΈ Architecture
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β Google Antigravity CLI (`agy`) β
βββββββββββββββ¬ββββββββββββββββββββββββββββββββ²βββββββββββββ
β β
[1] PreInvocation [2] PreToolUse
β β
βΌ βΌ
Auto-Recall Hook URI Guard Hook
(Injected as EphemeralMsg) (Block viking:// on fs tools)
β
βββββββββββββΊ OpenViking Daemon (http://127.0.0.1:1933)
β β²
[3] PostInvocation β
βΌ β
Auto-Capture Hook ββββββββββ€
(Incremental turns to session) β
β
[4] Stop β
βΌ β
Auto-Commit Hook βββββββββββ
(Session memory extraction)π Quick Start
Prerequisites
Node.js >= 18.0.0
Antigravity CLI (
agy) >= 1.2.0 (command -v agy)A running OpenViking server (defaults to
http://127.0.0.1:1933)
1. One-Click Installation
Clone the repository and run the installer:
git clone https://github.com/x-multibyte/agy-plugin-openviking.git
cd agy-plugin-openviking
./install.sh2. Manual Installation via agy
You can also install directly with the Antigravity CLI:
agy plugin validate /path/to/agy-plugin-openviking
agy plugin install /path/to/agy-plugin-openviking3. Verify Installation
Check installed components:
agy plugin listExpected output:
{
"imports": [
{
"name": "openviking",
"source": "antigravity",
"components": [
"skills",
"mcpServers",
"hooks"
]
}
]
}βοΈ Configuration
Credentials and endpoint resolution follow the standard OpenViking precedence hierarchy:
Environment Variables:
OPENVIKING_URL(default:http://127.0.0.1:1933)OPENVIKING_API_KEY(orOPENVIKING_BEARER_TOKEN)OPENVIKING_ACCOUNT/OPENVIKING_USER
CLI Configuration File:
~/.openviking/ovcli.confServer Configuration File:
~/.openviking/ov.conf
Low-Latency Mode
If response latency is critical, you can disable optional query expansion:
export OPENVIKING_RECALL_QUERY_EXPANSION=offOr configure in ~/.openviking/ovcli.conf:
{
"url": "http://127.0.0.1:1933",
"plugin": {
"recallQueryExpansion": "off"
}
}π§ͺ Testing
Run the included unit and conformance test suites:
npm testTests cover:
Antigravity
transcript_full.jsonllog parsing.JSON Schema and Antigravity plugin manifest conformance.
Lifecycle hook event definitions and syntax validation.
π License
This project is licensed under the GNU Affero General Public License v3.0 (AGPL-3.0) to align with OpenViking core components.
This server cannot be deployed
Maintenance
Related MCP Connectors
Persistent memory for AI agents. Search, store, and recall across sessions.
Persistent memory for AI agents. Search and store durable facts, preferences and decisions.
Universal persistent memory and knowledge retrieval layer for AI agents and LLMs.
Universal memory for AI agents and tools. Save, organize and search context anywhere.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceProvides AI agents with persistent, searchable memory that survives across conversations using semantic search, temporal versioning, and smart organization. Enables long-term context retention and cross-session continuity for AI assistants.14-
- AlicenseAqualityBmaintenanceProvides persistent, searchable memory across AI coding agent and chat history (Claude Code, Codex, Gemini CLI, ChatGPT, and more) via retrieval-augmented generation, enabling semantic and hybrid search to retain context across sessions.5MIT
- AlicenseAqualityDmaintenanceProvides persistent memory with semantic search for MCP-based AI agents, enabling them to store and recall information across sessions using vector embeddings.41MIT
- FlicenseNot gradedqualityBmaintenanceProvides a shared long-term memory across multiple AI clients, enabling persistent storage and retrieval of facts, preferences, decisions, and snippets with semantic search.-