claude-migrate-memory-to-doc

An agent skill by daymade, from daymade/claude-code-skills. Tags: documentation, local-workflow, migration, review, strategy.

What it does

Migrates Claude Code personal memory (the per-project memory/ directory) into tool-agnostic reference docs, so other AI CLIs that auto-load AGENTS.md (Codex primarily; the content architecture transfers to Cursor and others) working in the same directory can read the same user profile, collaboration preferences, and methodology instead of being blind to them. Use this whenever the user says things like "migrate my memory", "my memory is locked to Claude Code", "make Codex/Cursor read my profile", "memory should live in docs not one tool", or reports that a second AI tool doesn't know who they are; also use it when memory has grown bloated with content that should be shared across tools or projects. Covers diagnosis, the references/ + CLAUDE.md-inline + AGENTS.md-symlink architecture, multi-agent review, empirical codex verification, and memory cleanup. Inline only — it orchestrates review subagents and runs codex.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add daymade/claude-code-skills --skill claude-migrate-memory-to-doc

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/daymade/claude-code-skills
cp -r claude-code-skills/daymade-claude-code/claude-migrate-memory-to-doc ~/.claude/skills/claude-migrate-memory-to-doc

Safety box score

Grade D, elevated risk: 46 of 100 risk points, rated 2026-10-19. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill migrates Claude Code personal memory into `~/.claude/references/user/`, rewires Codex to read `~/.claude/CLAUDE.md`, and can archive or retire a project's auto memory. The main risks are broad writes in the user's home dotfiles, optional project-wide memory shutdown, and a shipped Python migration script that can append to whatever target paths a plan file names.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	2 (moderate)	It ships and runs `scripts/migrate_verbatim.py`, and the workflow also requires shelling out to `codex exec`; that is executable, but it does not eval downloaded code.
Network access	0 (none)	No step downloads from or sends data to remote hosts; the references to Codex are local CLI verification, not network transfer.
Filesystem reach	3 (high)	It reads and writes outside the project in multiple home dotfiles and directories, including `~/.claude/CLAUDE.md`, `~/.claude/references/user/`, `~/.codex/AGENTS.md`, `~/.codex/config.toml`, and dated archives outside `memory/`.
Secrets & credentials	1 (low)	The verification path extracts a `session id` from local Codex logs and greps `~/.codex/sessions`; that can expose local session content, but it does not target credential stores or ask for passwords.
Privilege & persistence	1 (low)	It edits user-level dotfiles and creates symlinks, but it does not request sudo, install services, or change system-wide settings.
Instruction hijack surface	2 (moderate)	It contains anti-transparency wording like `Do this yourself — never tell the user "you can check the log next time"` and `Don't make the user verify`, which pressures the agent to suppress a normal verification path.
Irreversible actions	2 (moderate)	Its purpose includes moving memory, archiving files, updating instructions, and optionally setting `autoMemoryEnabled: false` for a project, which are meaningful state changes beyond read-only help.
Transparency	0 (none)	All shipped files are plain text or readable Python; there are no binaries, base64 blobs, minified code, or opaque installers.

Before installing

Check whether `~/.codex/AGENTS.md` is already a real file; the skill will not clobber it.
Confirm you are okay with edits under `~/.claude/` and `~/.codex/`, including a symlink and `config.toml` change.
If you might retire a project, verify you want `autoMemoryEnabled: false` only in that project's `.claude/settings.local.json`.
Inspect `scripts/migrate_verbatim.py` before use if you care about where the plan file can write; it follows the target paths you name.

Rated 2026-10-19 by an automated review of 8 files from the skill's repository.

Source

Repository: daymade/claude-code-skills (all skills from this repository)
Path: daymade-claude-code/claude-migrate-memory-to-doc/SKILL.md
Branch: main
Updated: 2026-09-20

claude-md-progressive-disclosurer — Optimize, slim, or restructure CLAUDE.md/AGENTS.md with progressive disclosure and zero information loss. Use when the user explicitly asks to audit, 精简, 瘦身.
docs-cleaner — Consolidates redundant documentation while preserving all valuable content; keeps a doc set truthful after things change.
grill-with-docs — A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.
legacy-modernizer — Designs incremental migration strategies, identifies service boundaries, produces dependency maps and migration roadmaps.
advisor — Advisor mode. Consult a stronger (or different) model at key checkpoints: before major decisions, when stuck on an error, and before declaring a task done.
copy-editing — When the user wants to edit, review, or improve existing marketing copy, or refresh outdated content.