Skip to main content
command CX Codex

memory-migrate — Claude Code command: convert agent memory to YAML schema

memory-migrate — a command for Claude Code and Codex that converts legacy flat memory files (pitfalls.md, conventions.md, decisions.md) into a structured YAML schema with automatic entry classification.

MIT ★ 730 updated 2026-10-29

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add gmickel/flow-next
/plugin install flow-next@flow-next
Open on GitHub

What memory-migrate does

The command reads text files from .flow/memory/ and transforms each entry into categorized YAML format — the agent classifies content directly during migration. It supports autofix mode for fully automatic conversion and an optional file filter to migrate only pitfalls.md or a specific section.

After writing, the command verifies the output and can remove source files upon success. Pass arguments mode:autofix and optionally a filename (e.g., pitfalls.md) — the rest runs through the built-in flow-next-memory-migrate skill.

What it helps with

  • migrate agent memory from Markdown to YAML without losing entries
  • classify records by type automatically during the conversion
  • update .flow/memory/ structure to the new schema with verification
  • convert a single memory file while leaving others untouched

Example prompts for the agent

  • Migrate all agent memory: memory-migrate mode:autofix
  • Convert only pitfalls: memory-migrate mode:autofix pitfalls.md
  • Classify and transform decisions.md into YAML: memory-migrate decisions.md

How to install memory-migrate

1Plugin marketplace

/plugin marketplace add gmickel/flow-next
/plugin install flow-next@flow-next

2Manually: put the files in place

.claude/commands/memory-migrate.md

Call it with /memory-migrate. For all projects use ~/.claude/commands/

Questions about memory-migrate

How do I install memory-migrate in Claude Code?

Run /plugin marketplace add gmickel/flow-next, then /plugin install flow-next@flow-next. The command will be available as .claude/commands/memory-migrate.md.

What does memory-migrate do?

It converts legacy flat memory files (.flow/memory/pitfalls.md, conventions.md, decisions.md) into structured YAML format with automatic categorization for each entry.

Are there any prerequisites to run it?

You need the source files in .flow/memory/. The autofix mode runs fully automatic migration without additional input.

Source text

memory-migrate.md · MIT
---
name: memory-migrate
description: Migrate legacy `.flow/memory/{pitfalls,conventions,decisions}.md` flat files into the categorized YAML schema — host agent classifies each entry directly
argument-hint: "[mode:autofix] [optional scope hint — e.g. pitfalls.md]"
disable-model-invocation: true
---

# IMPORTANT: This command MUST invoke the skill `flow-next-memory-migrate`

The ONLY purpose of this command is to call the `flow-next-memory-migrate` skill. You MUST use that skill now.

**Arguments:** $ARGUMENTS

Pass the arguments to the skill verbatim. The skill handles mode detection (`mode:autofix` token), legacy-file detection, per-entry classification, categorized writes, verification, and optional cleanup.

© Gordon Mickel · MIT · gmickel/flow-next

Similar slash commands

All slash commands for Agent setup: memory, context, orchestration →

From the same set70

The whole set →

Project sponsors

Companies that keep this analytics open