vellum-memory-v2-migration

Migrate Vellum workspaces from memory v1 to memory v2 by backfilling concept pages and validating the wiki.

2|2|Updated May 13, 2026
One-click install
npx skills add https://github.com/autonomy-cloud/kairos-interface --skill vellum-memory-v2-migration
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: vellum-memory-v2-migration
Source: https://github.com/autonomy-cloud/kairos-interface/tree/main/skills/vellum-memory-v2-migration
Command: npx skills add https://github.com/autonomy-cloud/kairos-interface --skill vellum-memory-v2-migration

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you migrate an existing Vellum workspace from memory v1 to memory v2 by backfilling /workspace/memory/ concept pages, draining the buffer, validating the resulting wiki, and enabling the new retrieval format.

Core Features & Use Cases

  • Guided memory consolidation: Reads existing knowledge sources, plans a wiki taxonomy, writes concept pages, and establishes cross-linked structure for next-you retrieval.
  • Buffer distillation + archival: Routes durable facts onto concept pages and uses an idempotent helper to archive per-day entries while preserving unparsed items for review.
  • Safety guardrails + validation: Creates sentinel commits for traceability, pins to a high-quality model when available, runs assistant memory v2 validate, flips memory.v2.enabled only after passing validation, then re-embeds and refreshes activation.

Quick Start

Run vellum-memory-v2-migration when your workspace needs the first-time backfill to memory v2 and confirm the preflight prompt to proceed with the migration.

Frequently Asked Questions about vellum-memory-v2-migration

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I migrate my Vellum workspace from memory v1 to memory v2?

Migrate Vellum memory v1 to v2 by backfilling concept pages, draining the memory buffer, validating the wiki, and enabling memory.v2.enabled. This process distills existing knowledge sources into routable pages for next-you retrieval.

What is the safest way to backfill knowledge base concepts for personal assistants?

Safe knowledge base backfill involves reading wiki principles, generating sentinel commits for traceability, and pinning a high-quality inference session. Validation runs before the new memory format activates to ensure wiki integrity.

How does buffer distillation work during wiki concept page migration?

Buffer distillation routes durable facts onto concept pages and uses an idempotent helper to archive per-day entries. Unparsed items are preserved for review, ensuring no data is lost during the memory migration process.

Do I need to run embedding reindexing after enabling memory v2?

Embedding reindexing is required after enabling memory v2. You must run assistant memory v2 reembed and reembed-skills to refresh activation and ensure the new retrieval format routes queries to the correct concept pages.

When should I not use automated memory v2 migration?

Avoid automated memory v2 migration if your workspace lacks existing knowledge sources to distill, or if the preflight validation fails. The process requires passing assistant memory v2 validate before flipping the memory.v2.enabled flag.