MCPgee
MCPgee
MCPgeeMCPgee

Continuous Claude V3

v1.0.0•Knowledge & Memory•stable

Context management for Claude Code. Hooks maintain state via ledgers and handoffs. MCP execution without context pollution. Agent orchestration with isolated context windows.

agentsclaude-codeclaude-code-cliclaude-code-hooksclaude-code-mcp
Share:
3,781
Stars
0
Downloads
0
Weekly
0/5
View on GitHub

What is Continuous Claude V3?

Continuous Claude V3 is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to context management for claude code. hooks maintain state via ledgers and handoffs. mcp execution without context pollution. agent orchestration with isolated context windows.

Context management for Claude Code. Hooks maintain state via ledgers and handoffs. MCP execution without context pollution. Agent orchestration with isolated context windows.

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Context management for Claude Code. Hooks maintain state via

Use Cases

Manage context in Claude Code without pollution.
Orchestrate agents with isolated context windows and state ledgers.
parcadei

Maintainer

LicenseMIT
Languagepython
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux
View SourceBrowse All Servers

Installation

Manual Installation

npx continuous-claude-v3

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Continuous Claude V3

Continuous Claude v3 is a comprehensive context management and agent orchestration system for Claude Code that solves the problem of lost context across long sessions and agent handoffs. It uses PostgreSQL with pgvector for persistent memory, continuity ledgers (Markdown files tracking in-session progress), and structured YAML handoff files to transfer context between sessions without polluting the active context window. The system includes 109 skills, 32 specialized agents, and 30 lifecycle hooks that together enable autonomous multi-agent workflows with 95% token savings on code analysis via a 5-layer AST/CFG/DFG analysis stack.

Prerequisites

  • Python 3.11 or later installed
  • uv package manager installed (https://github.com/astral-sh/uv)
  • Docker installed and running (for the PostgreSQL + pgvector database)
  • Claude Code CLI installed and authenticated
  • Optional: Braintrust API key (session tracing), Perplexity API key (web search), NIA API key (documentation search)
1

Clone the repository

Clone the Continuous Claude repository to your local machine. The opc directory contains the setup wizard and all scripts.

git clone https://github.com/parcadei/Continuous-Claude-v3.git
cd Continuous-Claude-v3/opc
2

Run the setup wizard

The wizard performs 12 automated steps including Docker container setup, PostgreSQL schema initialization with pgvector, and installation of skills, agents, hooks, and rules into ~/.claude/.

uv run python -m scripts.setup.wizard
3

Configure optional API keys

Add optional API keys to your Claude Code settings for enhanced capabilities. Perplexity enables web search, Braintrust enables session analytics, and NIA enables documentation lookup.

# In ~/.claude/settings.json, add to the env section:
{
  "env": {
    "BRAINTRUST_API_KEY": "your-key",
    "PERPLEXITY_API_KEY": "your-key",
    "NIA_API_KEY": "your-key"
  }
}
4

Configure a remote PostgreSQL database (optional)

If you want to share memory across machines, set the CONTINUOUS_CLAUDE_DB_URL environment variable to point to a remote PostgreSQL instance with the pgvector extension. The pgvector schema is initialized from docker/init-schema.sql.

{
  "env": {
    "CONTINUOUS_CLAUDE_DB_URL": "postgresql://user:pass@hostname:5432/continuous_claude"
  }
}
5

Start a session with workflow routing

Open your project in Claude Code. Use the /workflow meta-skill to route to the appropriate Research, Plan, Build, or Fix workflow based on your current task.

/workflow
6

Save state and create handoffs between sessions

Use natural language phrases to trigger context saving. At the end of a session, say 'done for today' to create a handoff file. Resume the next day with 'resume work'.

Continuous Claude V3 Examples

Claude Desktop / Claude Code configuration

Continuous Claude v3 installs into Claude Code's ~/.claude/ directory rather than running as a standalone MCP server. The DATABASE_URL is set automatically by the wizard, but you can reference it in settings.

{
  "env": {
    "DATABASE_URL": "postgresql://localhost:5432/continuous_claude",
    "CLAUDE_OPC_DIR": "/path/to/Continuous-Claude-v3/opc",
    "CLAUDE_PROJECT_DIR": "/path/to/your/project"
  }
}

Prompts to try

Use these natural language commands inside a Claude Code session to activate the installed skills and agents.

- "Fix the login bug in auth.py" (activates /fix workflow, debug-agent, scout, and auto-creates a handoff)
- "Build a user dashboard" (activates /build workflow, plan-agent, architect, kraken)
- "Understand the codebase" (activates /explore deep, scout agent)
- "What could go wrong with this approach?" (activates /premortem with TIGERS/ELEPHANTS risk analysis)
- "Research authentication patterns" (activates oracle agent + perplexity-search)
- "Done for today" (triggers create_handoff to save full session context)

Troubleshooting Continuous Claude V3

Setup wizard fails at the Docker/PostgreSQL step

Ensure Docker Desktop is running before starting the wizard. The wizard starts a PostgreSQL container with the pgvector extension using the schema in docker/init-schema.sql. If the container already exists, run: docker rm -f continuous-claude-db and re-run the wizard.

Memory recall returns no results after storing learnings

The memory daemon analyzes thinking blocks at session end and stores embeddings asynchronously. Allow a few seconds after session end for processing. Manually trigger recall with: cd opc && uv run python scripts/core/recall_learnings.py --query 'your query'

Skills or hooks are not activating in Claude Code sessions

Confirm that ~/.claude/skills/ and ~/.claude/hooks/ contain the installed files. Restart Claude Code after installation. Check ~/.claude/settings.json to ensure hooks are registered. If symlink mode is desired instead of copy mode, manually remove the copied files and symlink them to the repo.

Frequently Asked Questions about Continuous Claude V3

What is Continuous Claude V3?

Continuous Claude V3 is a Model Context Protocol (MCP) server that context management for claude code. hooks maintain state via ledgers and handoffs. mcp execution without context pollution. agent orchestration with isolated context windows. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Continuous Claude V3?

Follow the installation instructions on the Continuous Claude V3 GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Continuous Claude V3?

Continuous Claude V3 works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Continuous Claude V3 free to use?

Yes, Continuous Claude V3 is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Learn More About MCP Servers

Getting Started with MCP

Set up your first MCP server in minutes

MCP Setup Guide

Configure MCP in Claude, Cursor & VS Code

All MCP Tutorials

18+ hands-on guides for developers

MCP FAQ

40+ answers about Model Context Protocol

Continuous Claude V3 Alternatives — Similar Knowledge & Memory Servers

Looking for alternatives to Continuous Claude V3? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.

MemPalace

★ 52.6k

A local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically

Kratos

★ 25.7k

🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.

Context Mode

★ 15.4k

An MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm

Memu

★ 13.7k

Memory for 24/7 proactive agents like OpenClaw.

MemOS

★ 9.3k

MemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u

Everos

★ 5.4k

Build, evaluate, and integrate long-term memory for self-evolving agents.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Knowledge & MemoryBrowse All Servers

Quick Config Preview

{ "mcpServers": { "continuous-claude-v3": { "command": "npx", "args": ["-y", "continuous-claude-v3"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Continuous Claude V3?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides
Explore All ServersRead Our Guides