openclaw/docs/install/migrating-claude.md

summary, read_when, title

summary	read_when	title
Move Claude Code and Claude Desktop local state into OpenClaw with a previewed import	You are coming from Claude Code or Claude Desktop and want to keep instructions, MCP servers, and skills You need to understand what OpenClaw imports automatically and what stays archive-only	Migrating from Claude

OpenClaw imports local Claude state through the bundled Claude migration provider. The provider previews every item before changing state, redacts secrets in plans and reports, and creates a verified backup before apply.

Onboarding imports require a fresh OpenClaw setup. If you already have local OpenClaw state, reset config, credentials, sessions, and the workspace first, or use `openclaw migrate` directly with `--overwrite` after reviewing the plan.

Two ways to import

The wizard offers Claude when it detects local Claude state.

```bash
openclaw onboard --flow import
```

Or point at a specific source:

```bash
openclaw onboard --import-from claude --import-source ~/.claude
```

Use `openclaw migrate` for scripted or repeatable runs. See [`openclaw migrate`](/cli/migrate) for the full reference.

```bash
openclaw migrate claude --dry-run
openclaw migrate apply claude --yes
```

Add `--from <path>` to import a specific Claude Code home or project root.

What gets imported

- Project `CLAUDE.md` and `.claude/CLAUDE.md` content is copied or appended into the OpenClaw agent workspace `AGENTS.md`. - User `~/.claude/CLAUDE.md` content is appended into workspace `USER.md`. MCP server definitions are imported from project `.mcp.json`, Claude Code `~/.claude.json`, and Claude Desktop `claude_desktop_config.json` when present. - Claude skills with a `SKILL.md` file are copied into the OpenClaw workspace skills directory. - Claude command Markdown files under `.claude/commands/` or `~/.claude/commands/` are converted into OpenClaw skills with `disable-model-invocation: true`.

What stays archive-only

The provider copies these into the migration report for manual review, but does not load them into live OpenClaw config:

• Claude hooks
• Claude permissions and broad tool allowlists
• Claude environment defaults
• CLAUDE.local.md
• .claude/rules/
• Claude subagents under .claude/agents/ or ~/.claude/agents/
• Claude Code caches, plans, and project history directories
• Claude Desktop extensions and OS-stored credentials

OpenClaw refuses to execute hooks, trust permission allowlists, or decode opaque OAuth and Desktop credential state automatically. Move what you need by hand after reviewing the archive.

Source selection

Without --from, OpenClaw inspects the default Claude Code home at ~/.claude, the sampled Claude Code ~/.claude.json state file, and the Claude Desktop MCP config on macOS.

When --from points at a project root, OpenClaw imports only that project's Claude files such as CLAUDE.md, .claude/settings.json, .claude/commands/, .claude/skills/, and .mcp.json. It does not read your global Claude home during a project-root import.

Recommended flow

```bash openclaw migrate claude --dry-run ```

The plan lists everything that will change, including conflicts, skipped items, and sensitive values redacted from nested MCP `env` or `headers` fields.

```bash openclaw migrate apply claude --yes ```

OpenClaw creates and verifies a backup before applying.

```bash openclaw doctor ```

[Doctor](/gateway/doctor) checks for config or state issues after the import.

```bash openclaw gateway restart openclaw status ```

Confirm the gateway is healthy and your imported instructions, MCP servers, and skills are loaded.

Conflict handling

Apply refuses to continue when the plan reports conflicts (a file or config value already exists at the target).

Rerun with `--overwrite` only when replacing the existing target is intentional. Providers may still write item-level backups for overwritten files in the migration report directory.

For a fresh OpenClaw install, conflicts are unusual. They typically appear when you re-run the import on a setup that already has user edits.

JSON output for automation

openclaw migrate claude --dry-run --json
openclaw migrate apply claude --json --yes

With --json and no --yes, apply prints the plan and does not mutate state. This is the safest mode for CI and shared scripts.

Troubleshooting

Pass `--from /actual/path` (CLI) or `--import-source /actual/path` (onboarding). Onboarding imports require a fresh setup. Either reset state and re-onboard, or use `openclaw migrate apply claude` directly, which supports `--overwrite` and explicit backup control. Claude Desktop reads `claude_desktop_config.json` from a platform-specific path. Point `--from` at that file's directory if OpenClaw did not detect it automatically. By design. Claude commands are user-triggered, so OpenClaw imports them as skills with `disable-model-invocation: true`. Edit each skill's frontmatter if you want the agent to invoke them automatically.

Related

• openclaw migrate: full CLI reference, plugin contract, and JSON shapes.
• Migration guide: all migration paths.
• Migrating from Hermes: the other cross-system import path.
• Onboarding: wizard flow and non-interactive flags.
• Doctor: post-migration health check.
• Agent workspace: where AGENTS.md, USER.md, and skills live.