> ## Documentation Index
> Fetch the complete documentation index at: https://docs.case.dev/llms.txt
> Use this file to discover all available pages before exploring further.
# Clio Manage
> Matter and folder import, export, and selectable collections
`provider: "clio"` · Import and export · Generally available
The Clio connector browses matters and their document folders, mirrors a matter root or a nested folder into a vault, and can export vault work product back to Clio.
US region only. Clio Grow and non-US regions are not supported, and the scope tier is always `clio.us`.
## Browsing
Browse with no parameters to list the matters the connected user can reach. Each matter carries a folder reference containing both its matter id and root folder id, so you can import a whole matter or descend into a subfolder first.
```bash title="cURL" theme={"theme":{"light":"github-light","dark":"one-dark-pro"}}
curl "https://api.case.dev/connectors/v1/connections/conn_abc123/browse" \
-H "Authorization: Bearer $CASEDEV_API_KEY"
```
```json title="Response" theme={"theme":{"light":"github-light","dark":"one-dark-pro"}}
{
"items": [
{
"id": "987654",
"name": "Smith v. Hospital",
"kind": "matter",
"browse_ref": { "kind": "folder", "folder_id": "987654", "container_id": "123456" }
}
]
}
```
Matter and folder ids are stable across renames and moves. Display names and paths never participate in link identity, so a user reorganizing their Clio folders does not break an existing link.
Clio's request budget is **45 requests per minute**, and interactive browsing shares it with background runs.
Page browse results rather than eagerly walking the tree, and expect `429` with `Retry-After` while a sync is running for the same account. Honor it instead of retrying immediately.
## Selectable collections
Clio is the only provider that exposes more than one collection. Choose with `remote.resource_type`:
| `resource_type` | Direction | Contents |
| ---------------- | ----------------- | ----------------------------------------------------------- |
| `documents` | Import and export | Matter documents. The default when omitted. |
| `communications` | Import only | Matter communications, imported as searchable vault content |
| `tasks` | Import only | Matter tasks, imported as searchable vault content |
Each link selects exactly one collection, so cursors and failures stay isolated. To bring documents and communications into the same vault, create two links.
Asking for an export link on `communications` or `tasks` returns `400`.
## Importing a matter
```bash title="cURL" theme={"theme":{"light":"github-light","dark":"one-dark-pro"}}
curl -X POST https://api.case.dev/connectors/v1/transfer \
-H "Authorization: Bearer $CASEDEV_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"direction": "import",
"connection_id": "conn_abc123",
"remote": {
"container_id": "123456",
"folder_id": "987654",
"path": "Smith v. Hospital"
},
"vault_id": "vault_abc123"
}'
```
## Change detection
Clio reports changes by modification time rather than a native delta feed.
* The first run captures a timestamp, then recursively lists the selected root
* Later runs query documents updated since that timestamp, with a short overlap for safety
* A scheduled daily full reconcile catches deletions, newly inaccessible documents, and files moved out of the selected subtree—things a timestamp filter cannot report
Content identity is the document's latest version, not the document record. A rename or folder move updates path metadata **without re-downloading the file**.
## Export and two-way sync
Clio supports `direction: "export"` and `direction: "both"`.
With `both`, you get a paired import link and export link sharing a `pair_id`. Unless you pass `export_destination`, exports land in a `CaseMark Output` folder created under the selected Clio folder.
### Lane ownership
Two-way sync could easily become a loop. It does not, because each side owns a lane:
* **Clio-native documents** flow into the vault
* **Vault-native documents** flow out to Clio
Only vault-native objects are exported—documents that arrived *from* Clio are excluded, so nothing round-trips. Documents created by the export carry provenance markers, and Clio change discovery skips them, so an exported file is never re-imported as a duplicate.
Edits made to a Case.dev-owned document inside `CaseMark Output` do not flow back into the vault.
If both sides changed since the last successful export, the item **halts with a conflict** rather than silently overwriting either version. Surface these to the user; the connector will not guess.
### Collision policy
| `policy.collisions` | Behavior on a name collision |
| ------------------- | -------------------------------------------- |
| `version` (default) | Creates a new sibling document |
| `overwrite` | Adds a new version to the colliding document |
| `skip` | Leaves the existing Clio document untouched |
Export never deletes Clio documents, whatever the delete policy says.
## Reconnection
Clio refresh tokens are long-lived, so reconnection is rare.
One exception: Clio does not upgrade permissions on an existing OAuth grant when the application's permissions change. Accounts that authorized before export was enabled **must reconnect** before they can export.
A write-side `403` moves the connection to `reauth_required`. Prompt the user to reconnect rather than treating it as a permanent failure.
## Capabilities at a glance
| | |
| ---------------- | -------------------------------------------- |
| Containers | Matters |
| Browse depth | Folder |
| Change detection | Updated-since, with scheduled full reconcile |
| Export | Yes |
| Collections | `documents`, `communications`, `tasks` |
| Scope tier | `clio.us` |
| Rate budget | 45 requests/min, 4 concurrent transfers |
| Webhooks | Not supported |
## Next steps
Link mechanics, policies, and progress
Throttling, reauthorization, and conflicts
Compare against the other supported providers