Lov Cc Mv logo

Lov Cc Mv

Organization
lovstudio
lov-cc-mv

Move a project folder AND migrate all its Claude Code state in one shot — session store, prompt-up-arrow history, running-session records. Supports both directory-level moves AND session-level cherry-picking (by id, regex on first user prompt, or interactive picker). Use whenever the user wants to rename/move a project directory and keep `claude --resume` working, or wants to move a subset of chats from one project to another. 移动/重命名项目目录并迁移所有 CC 历史,或按 session 粒度把某个话题的对话搬到另一个项目。

Overview

Publisherlovstudio
Repositoryskills
Skill namelov-cc-mv
Stars
67
Forks
17
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by lovstudio on GitHub. Read the source before you install it.

Installation

Install the Lov Cc Mv AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/lovstudio/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/skills/cc-migrate-session .claude/skills/lov-cc-mv
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Lov Cc Mv in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Lov Cc Mv on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Lov Cc Mv is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

会话搬家 · Session Mover

Two modes in one tool:

Directory-level (default) — four things in one shot:

  1. mv FROM TO on disk (fs.renameSync — instant, preserves everything)
  2. Rewrites ~/.claude/projects/<slug>/*.jsonl session store — including every sub-directory slug
  3. Rewrites ~/.claude/history.jsonl (prompt up-arrow recall)
  4. Rewrites ~/.claude/sessions/*.json (running-session records)

Session-level (opt-in, via --session/--grep/--pick) — migrate a subset of chats:

  1. Copies only the selected .jsonl files from FROM's slug dir into TO's
  2. Rewrites each file's cwd from FROM → TO
  3. Rewrites only the matching running-session records
  4. Does not touch history.jsonl or fs mv (project itself isn't being moved)
  5. Sub-dir slugs are ignored (root only by design)

Old slug dirs and source sessions are left intact unless --delete-source is passed.

When to Trigger

YES — directory-level when:

  • User wants to move/rename a project folder (prospective move — we do the mv)
  • User already moved the folder externally and CC lost history (post-move recovery — use --no-mv or the cc-migrate-session alias)
  • Sub-dir sessions under FROM should come along (handled automatically)

YES — session-level when:

  • User wants to migrate only some sessions from FROM to TO — identified by id, topic/content, or to be picked interactively
  • User said "只搬这几个" / "those chats about X" / "not all of them"
  • User needs to split one project's chat history across two projects

NO — don't invoke when:

  • Renaming a file, function, variable, or branch (not the project root)
  • General question about CC's storage model (explain, don't migrate)
  • Paths are ambiguous — ask first

Workflow

Step 1 — Gather FROM and TO

User saidFROMTO
"把 /a 搬到 /b" / "mv /a to /b"/a/b
"rename ~/foo to ~/bar"~/foo~/bar
"this project used to be at /old" (cwd is the new location)/oldprocess.cwd()
"本项目已迁移到 /new" (cwd is the old location)process.cwd()/new

If either side is ambiguous, ask once with AskUserQuestion. Don't guess.

Always expand ~ and resolve to absolute paths before running the CLI.

Step 2 — Decide directory-level vs session-level

Directory-level if:

  • User said "move the whole project" / "rename the folder"
  • User wants fs mv (the folder itself is moving on disk)
  • User wants ALL sessions (including any sub-dir sessions) migrated

Session-level if:

  • User said "only the sessions about X" / "just these chats" / "not all of them"
  • User referenced sessions by content/topic or explicit id
  • The folder itself should stay put (only CC state for a subset is moving)

Step 3 — Session-level: resolve the session set

The key design point: AI should auto-filter whenever possible, only prompt the user when ambiguous.

  1. Run:

    bash
    npx -y @lovstudio/cc-mv <FROM> --list-sessions --json

    Parse sessions[*] — each has sessionId, firstUserPrompt, mtime, sizeBytes, messageCount.

  2. Match the user's intent against firstUserPrompt yourself:

    • If the topic is unambiguous (e.g. "about command vs skill" + exactly one session's prompt obviously matches) → pick it directly, proceed to Step 4 with --session <id> flags.
    • If a regex captures it cleanly (e.g. topic = "slash command") → use --grep '<pattern>' instead of passing ids.
    • If multiple sessions could match and you can't disambiguate from firstUserPrompt alone → show the user the shortlist and ask which ones.
    • If the user asked for interactive picking explicitly → run with --pick and let the CLI do the UX.
  3. Exclude the current running CC session from the selection — it's the one the user is talking to you from right now, migrating it mid-conversation will break things. (Its sessionId matches the session file that was written to most recently; safer signal: ask the user to confirm if ambiguous.)

Step 4a — Dry-run + json to preview (directory-level)

bash
npx -y @lovstudio/cc-mv <FROM> <TO> --dry-run --json

Parse the JSON. Tell the user:

  • Total sessions and affected slug count (pairs[*] with sessionCount > 0)
  • If pairs.length > 1: flag that sub-directories also have CC history
  • If toDirExistsOnDisk and FROM also exists: warn — CLI will refuse the fs mv
  • If any pairs[i].toSlugDirExists: warn — destination slug dir will be merged

If totalSessions === 0 AND the user wanted post-move recovery (FROM path doesn't exist): stop, tell them either (a) FROM path is wrong, or (b) CC never ran there.

Step 4b — Dry-run + json to preview (session-level)

bash
npx -y @lovstudio/cc-mv <FROM> <TO> --session <id> --session <id> --dry-run --json
# or
npx -y @lovstudio/cc-mv <FROM> <TO> --grep '<pattern>' --dry-run --json

The JSON's sessionLevel: true, resolvedSessionIds: [...], and pairs[0].sessionFilter: [...] confirm which sessions will move.

Summarize for the user: "Will migrate N session(s): ". Get confirmation before executing.

Step 5 — Confirm

For directory-level with sub-dirs: ask "Found N sub-dir(s) with CC history. Migrate everything? [Y/n]" — default yes.

For session-level: always summarize the selected sessions (by first-prompt excerpt) and ask "Migrate these N session(s)?" — especially if you resolved them from a regex or topic match, so the user can catch false positives.

Also ask about --delete-source if the user said anything like "move" (vs "copy") — default to keeping source as safety net, only pass --delete-source when the user explicitly wants it.

Step 6 — Execute

Directory-level:

bash
npx -y @lovstudio/cc-mv <FROM> <TO> --yes --json

For post-move recovery (FROM already moved externally):

bash
npx -y @lovstudio/cc-mv <FROM> <TO> --yes --no-mv --json
# OR equivalently:
npx -y @lovstudio/cc-migrate-session <FROM> <TO> --yes --json

Session-level:

bash
npx -y @lovstudio/cc-mv <FROM> <TO> --session <id1> --session <id2> --yes --json
# or with regex:
npx -y @lovstudio/cc-mv <FROM> <TO> --grep '<pattern>' --yes --json
# optionally add --delete-source

Parse phase: "done":

  • result.slugsMigrated, result.jsonlFilesWritten, result.cwdRewrites
  • result.historyRewrites (0 in session-level mode — expected)
  • result.runningSessionRewrites
  • result.sourceSessionsDeleted (only non-zero with --delete-source)
  • fsMvMethod: "rename", "shell-mv", or null (session-level / --no-mv)
  • restartHint.cd + restartHint.command

Step 7 — Tell user to restart CC

Directory-level:

✓ Moved FROM → TO and migrated N session(s) across M slug dir(s).
✓ Also rewrote prompt history and running-session records.

Restart Claude Code in the new location:

  cd <TO>
  claude --resume

(The old slug dirs at ~/.claude/projects/<old-slug>* are untouched — delete
them once you've verified --resume works.)

Session-level:

✓ Migrated N session(s) from FROM to TO.
✓ Source sessions {kept as safety net | deleted}.

To resume one of the migrated sessions:

  cd <TO>
  claude --resume <session-id>

IMPORTANT: The CURRENT Claude Code session cannot "switch" its own cwd mid-session. The user must exit and re-invoke claude from the new directory. State this clearly.

CLI Reference

npx -y @lovstudio/cc-mv <FROM> <TO> [options] npx -y @lovstudio/cc-mv <FROM> [<TO>] --list-sessions [--json]

OptionPurpose
-y, --yesSkip confirmation prompt
--dry-runShow plan, don't write
--no-mvSkip the filesystem mv (only migrate CC state — post-move recovery)
--jsonMachine-readable output (use this from the skill)
--projects-dir <dir>Override CC projects dir (default ~/.claude/projects)
--session <id>Session-level: migrate only this id (repeatable)
--grep <pattern>Session-level: migrate sessions whose first user prompt matches the regex (case-insensitive)
--pickSession-level: interactive numbered picker
--list-sessionsPrint session summaries and exit (pair with --json for scripting)
--delete-sourceDelete migrated source sessions after copy+rewrite (default keeps them)

Backwards-compatible alias

npx -y @lovstudio/cc-migrate-session <FROM> <TO> — same tool, but defaults to --no-mv (only migrates CC state, doesn't touch the filesystem). Use when the user already moved the folder externally.

Sub-directory Discovery (directory-level only)

CC's slug rule is: replace every non-[A-Za-z0-9] with -. So FROM/sub slugifies to <fromSlug>-<subSlug>.

The CLI lists ~/.claude/projects/ and takes every slug matching slug === fromSlug || slug.startsWith(fromSlug + "-"). That catches FROM and all descendants in one readdir. It then reads each slug's first jsonl to recover the original absolute sub-path (since slug → path isn't reversible), builds the migration pair, and proceeds.

Session-level mode deliberately ignores sub-dir slugs — the user is cherry-picking sessions, not moving the whole project.

Safety

  • Old slug dirs are never deleted in directory-level mode. Copy-then-rewrite. Old state survives.
  • Source session files are kept by default in session-level mode. Pass --delete-source explicitly to remove them.
  • fs mv refuses if TO already exists on disk (avoid overwrite).
  • Slug-dir merging is default when dest slug dir exists — conflicting jsonls overwritten.
  • Session-level + --mv is rejected (doesn't make sense — the folder itself isn't moving).
  • Malformed jsonl lines are passed through unchanged.

Tell the user to verify claude --resume works at TO before rm -rf of old slug dirs.

通用反馈闭环

用户在 Skill 驱动任务中提出修改意见时,继续当前产物前必须执行:

  1. 先判断意见是 task-specific(仅本次)还是 reusable(可跨任务复用)。
  2. task-specific 只修改当前任务,不改 Skill。
  3. reusable 先确定作用域:领域规则先更新对应 canonical Skill;适用于所有 Skill 的规则先更新共享规范。
  4. 完成规则更新、版本、lint 与分发核验后,再把修改应用到当前任务。
  5. reusable 修改会使此前的“确认”“继续”“发吧”失效;完成当前产物修改和回读后必须停下,等待用户下一步指示,不自动进入发布、提交或其他外部写入。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Lov Cc Mv AI skill do?

Move a project folder AND migrate all its Claude Code state in one shot — session store, prompt-up-arrow history, running-session records. Supports both directory-level moves AND session-level cherry-picking (by id, regex on first user prompt, or interactive picker). Use whenever the user wants to rename/move a project directory and keep `claude --resume` working, or wants to move a subset of chats from one project to another. 移动/重命名项目目录并迁移所有 CC 历史,或按 session 粒度把某个话题的对话搬到另一个项目。

Why use Lov Cc Mv on TypingMind?

Because you install it once and use it with any model. Lov Cc Mv is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Lov Cc Mv in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/lovstudio/skills/tree/main/skills/cc-migrate-session. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Lov Cc Mv?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Lov Cc Mv?

As many as you like. As long as a model supports skills, you can use Lov Cc Mv with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Lov Cc Mv AI skill free?

Yes. It is published on GitHub by lovstudio under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇