A high-velocity research orchestration engine designed to manage persistent state, synthesis, and autonomous verification for agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install researchbrain
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install researchbrain 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 specialized state manager designed for autonomous agentic research within the Openclaw Skills ecosystem. It provides a robust local SQLite infrastructure to store research artifacts, findings, and links, ensuring that agents maintain a persistent memory across sessions. By enabling divergent reasoning through hypothesis branching and active verification, it allows developers to build sophisticated research agents that can self-correct and synthesize information using local embeddings.
This skill is built for high-velocity research environments where data integrity and cross-agent collaboration are paramount. By leveraging a native Model Context Protocol (MCP) server, researchvault enables multiple agents to interact with a shared research state, ensuring that complex discoveries are properly indexed, verified, and ready for synthesis.
researchvault Use Cases
- Tracking complex, multi-step research trajectories with persistent state management.
- Exploring parallel research paths using divergent reasoning and hypothesis branching.
- Automating the verification of low-confidence data through self-correcting agent missions.
- Integrating cross-agent collaboration via the built-in MCP server for shared context.
- Implementing continuous background monitoring of external sources and URLs using watchdog mode.
How researchvault Works
- Initialize a project vault with a specific objective and unique identifier to track the lifecycle of the research.
- Ingest raw data from multiple sources, such as Reddit or web URLs, directly into the local SQLite persistence layer.
- Execute the synthesis engine to automatically discover and map links between disparate findings using local embeddings.
- Run active verification missions to plan and execute checks on low-confidence data, allowing agents to self-correct.
- Expose the research state via the MCP server to allow other agents in the Openclaw Skills ecosystem to query and contribute to the vault.
researchvault Setup
To get started with this skill, ensure you have Python 3.13 and the uv package manager installed. Use the following commands to initialize the environment:
# Create a virtual environment and install the skill in editable mode
uv venv
uv pip install -e .
Once installed, you can initialize your first research project using the CLI:
uv run python scripts/vault.py init --id "your-project-id" --name "Project Name" --objective "Define your goal"
researchvault Data Schema & Taxonomy
The researchvault skill organizes data within a structured SQLite database to ensure high-velocity access and persistence. The schema includes:
| Table | Description |
|---|---|
| Artifacts | Stores raw documents, source text, and ingested files. |
| Findings | Individual units of verified information or data points. |
| Links | Relationship mappings generated by the synthesis engine. |
| Branches | Parallel research paths and hypotheses for divergent reasoning. |
| Verification | Logs for self-correction missions and data confidence scoring. |
researchvault Advanced Features
- Native MCP Server support for standardized cross-agent communication.
- Watchdog Mode for continuous, autonomous background monitoring of web queries.
- Synthesis Engine that utilizes local embeddings for automated relationship discovery between artifacts.
- Multi-source ingestion pipelines (scuttle) for rapid data collection from social platforms and web sources.
- Divergent reasoning framework for managing complex, branching research hypotheses without state collision.
SKILL.md
Loading
Related Openclaw Skills
ResearchVault
A sophisticated state management and orchestration engine designed to empower autonomous agents with persistent memory, synthesis capabilities, and verification workflows.
ResearchVault
An autonomous state manager and orchestration engine for persistent, high-velocity AI research missions.
ResearchVault
An autonomous state manager for agentic research that handles persistent storage, reasoning branches, and automated verification.
ResearchVault
ResearchVault is a high-velocity research orchestration engine that manages persistent state, synthesis, and autonomous verification for AI agents.
Vault
A local orchestration engine for managing long-running research tasks with high reliability and zero external costs.
ResearchVault
A local-first research orchestration engine designed to manage persistent state, synthesis, and autonomous verification for AI agents.
Featured*
