ResearchVault for Openclaw

A local-first orchestration engine that manages persistent research state, synthesis, and autonomous verification for AI agents.

lraivisto
v3.0.5
Feb 16, 2026
0
0
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install researchvault

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install researchvault using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is ResearchVault?

ResearchVault is a robust, privacy-centric research orchestration engine designed to handle complex data synthesis and state management. As a specialized tool often used alongside Openclaw Skills, it prioritizes local storage and security, keeping all research data in a local SQLite database rather than the cloud.

The engine is built for developers and researchers who require strict control over their network transparency and data provenance. It features hardened SSRF protection, ensuring that outbound connections only occur for authorized search tasks. By providing a structured way to manage research objectives and ingested data, ResearchVault enables AI agents to perform deeper, more reliable analysis without sacrificing privacy.

ResearchVault Use Cases

  • Orchestrating multi-step research projects with persistent state management.
  • Managing autonomous data synthesis for trend analysis or technical deep-dives.
  • Creating a secure, local-first repository for sensitive research data.
  • Automating data ingestion from web sources or local file systems using watchdog services.
  • Integrating research capabilities into AI agents via the Model Context Protocol (MCP).

How ResearchVault Works

  1. Project Initialization: Use the CLI to define a research objective and create a named vault instance.
  2. Data Ingestion: Scuttle web sources or local content into the vault, which is then stored in a local SQLite database.
  3. Strategy Execution: Run the autonomous strategist to analyze ingested data and generate synthesized insights.
  4. State Persistence: All progress, metadata, and findings are saved locally, allowing for long-running research sessions.
  5. Interface Access: Optionally launch the web-based Portal or MCP server to interact with the research data through a UI or AI agent.

ResearchVault Setup

To get started with ResearchVault as part of your Openclaw Skills toolkit, follow these installation steps:

# Create and activate a virtual environment
python3 -m venv .venv
source .venv/bin/activate

# Install the package in editable mode
pip install -e .

# Initialize a new research project
python scripts/vault.py init --objective "Your Research Goal" --name "Project-Name"

ResearchVault Data Schema & Taxonomy

ResearchVault uses a structured approach to manage research data locally. The following table outlines the primary data organization:

Component Location / Method Description
Core Database ~/.researchvault/research_vault.db SQLite database containing all research state and ingested content.
Portal State ~/.researchvault/portal Directory for managing portal-specific metadata and sessions.
Metadata Internal Schema Tracks project IDs, objectives, and timestamps for synthesis.
Auth Secrets .portal_auth Stores generated tokens for secure web access.

ResearchVault Advanced Features

  • MCP Server Integration: Exposes research capabilities to any agent supporting the Model Context Protocol.
  • Directory Watchdog: Automatically ingests new files from specified directories for real-time research updates.
  • SSRF Hardening: Advanced security layer that blocks internal network access by default to prevent data leakage.
  • Web Portal: A dedicated frontend for visualizing research progress, accessible via tokenized URLs.
  • Multi-Provider Search: Built-in support for Brave Search, Serper, and SearXNG for comprehensive data gathering.

SKILL.md


Loading

Related Openclaw Skills

Featured*