> For the complete documentation index, see [llms.txt](https://evo-5.gitbook.io/evo/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://evo-5.gitbook.io/evo/concepts/workspace-import.md).

# Workspace Import

When a session opens in a project directory, evo imports existing agent memory files into the project scope. This imported knowledge is then recalled like any other memory.

Skill files are **not** imported. A `SKILL.md` reaches the model as a catalog entry (name, trigger, path) discovered from disk at recall time — see [Skills vs Memories](/evo/concepts/skills-vs-memories.md). Copying its body into a memory row would duplicate what is already on disk, and memories render inline while skills render as one line.

## What Gets Imported

| File / Directory                                                                                | Memory Kind |
| ----------------------------------------------------------------------------------------------- | ----------- |
| `CLAUDE.md`, `.claude/CLAUDE.md`                                                                | fact        |
| `AGENTS.md`, `agents.md`, `.agent/AGENTS.md`                                                    | constraint  |
| `.paper/AGENT_MEMORY.md`, `.paper/**/*.md`                                                      | fact        |
| `.claude/commands/**/*.md`, `.claude/agents/**/*.md`                                            | procedure   |
| `.codex/**/*.md`, `.copilot/instructions/**/*.md`, `.copilot/prompts/**/*.md`, `.agent/**/*.md` | constraint  |

## How Import Works

Each imported file becomes one memory item with:

* **Title**: Path relative to the workspace root
* **Tags**: `workspace-import` plus `tool:<tool>` (e.g., `tool:claude`)
* **Source**: `runtime: 'workspace-import'` and the absolute file path

### What's Skipped

* YAML frontmatter is stripped
* `SKILL.md` files, wherever they sit — they belong to the disk-skill catalog
* Everything under `.paper/agents/skills/` — evo's own skills, owned by the skills table
* Cache directories: `.tmp/`, `node_modules/`, `.git/`
* `*.memory.md` skill-experience files are skipped
* Empty documents are skipped

### Idempotent Updates

Import is idempotent:

* Items are upserted by `(project scope, title)`
* Changed files update existing items in place
* Rows whose file has disappeared are pruned — an import row is a projection of a file, and a projection of nothing is noise

Pruning keys on `source.runtime`, never on the `workspace-import` tag: a memory evo distilled *about* importing can carry that tag and has no file behind it.

A project is imported once on first session start. To force a re-import:

```ts
await ctx.evo.importWorkspace('/workspace/app', { force: true })
```

Or via HTTP API:

```bash
curl -X POST http://localhost:3000/evo/import-workspace \
  -H "Content-Type: application/json" \
  -d '{"cwd": "/workspace/app", "force": true}'
```

## Why This Design

Imported workspace files are a **projection of what exists on disk**. evo treats them as read-only references:

* They're recalled alongside evo's own memories
* They're never modified by reflection
* They're never evicted by capacity limits

This ensures your hand-written rules and procedures remain authoritative. One reflection can't accidentally delete your `CLAUDE.md` guidance.

## Disabling Import

In DeepSeek Harness, set in plugin config:

```yaml
- id: evo-deepseek
  name: evo/deepseek
  config:
    workspaceImport: false
```

In Claude Code / Codex, set environment variable:

```bash
export EVO_HOOK_IMPORT=0
```


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://evo-5.gitbook.io/evo/concepts/workspace-import.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.