Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
ResearchVault is a local-first research orchestration engine that stores all data in a local SQLite database. It manages persistent state, synthesis, and autonomous verification for agents, with optional background services including an MCP server and a watchdog process.
All data stays on-device in a local SQLite file with no cloud sync, making it suitable for sensitive research where external data exposure is a concern.
When to use it
- Tracking AI industry trends across multiple web sources
- Building a sourced literature review from ingested URLs
- Running autonomous research strategies on a named project
- Managing long-running research sessions with persistent state
- Accessing collected research through a local web portal
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: research AI trends for 2026 and synthesize findings
- 1Initialize a new project with `vault.py init --objective 'Analyze AI trends' --name 'Trends-2026'`
- 2Ingest relevant URLs using `vault.py scuttle <url> --id trends-2026`
- 3Run the autonomous strategist with `vault.py strategy --id trends-2026`
- 4Review synthesized output from the local SQLite database
A structured research synthesis with sourced findings stored locally under the Trends-2026 project
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.