Summary
The release lets teams create and manage reusable skills for Lightdash agents, invoke them through slash commands, and expose them through the platform’s MCP server. It also adds scoped management controls and records lasting user corrections.
Why it matters
This turns organisation-specific analytics procedures into governed, reusable agent capabilities rather than one-off prompts.
Related coverage
- Docker publishes an MCP Gateway OAuth helper
- Claude Code 2.1.282 hardens telemetry, permissions and session recovery
- Payload CMS adds agent-assisted upgrades in its 4.0 canary