Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
v0.2.0 release: M2M relationships, cascading ops, simplified CLI, MCP link/unlink tools
What This Skill Does
KameleonDB is a schema-flexible database for agents that stores and queries structured data without upfront schema design. The schema evolves automatically as new fields are discovered, and old records remain intact. Supports SQLite for local use and PostgreSQL for production.
Unlike traditional relational databases, it removes the need for migration planning, letting agents start storing data immediately and restructure as patterns emerge.
When to use it
- Tracking contacts and companies across agent sessions
- Building a knowledge base from web scraping or API ingestion
- Managing tasks where requirements change over time
- Storing API response data with unknown or variable structure
- Creating a lightweight CRM to track leads, deals, or tickets
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: track the people I meet and add new details as I learn them
- 1Run `kameleondb --json schema create Contact --field 'name:string:required' --field 'email:string:unique'` to create the entity
- 2Run `kameleondb --json data insert Contact '{"name":"Alice","email":"alice@example.com"}'` to add a record
- 3Run `kameleondb --json schema alter Contact --add 'linkedin_url:string' --reason 'Found LinkedIn profiles for contacts'` when a new field is needed
- 4Run `kameleondb --json data update Contact <id> '{"linkedin_url":"https://linkedin.com/in/alice"}'` to fill in the new field
- 5Run `kameleondb --json schema list` to review all tracked entities
Contact records stored and queryable via SQL, with schema updated in place and no existing records broken
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.