> ## Documentation Index
> Fetch the complete documentation index at: https://docs.crewship.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Consolidate Commands

> Force an immediate memory consolidation run.

# crewship consolidate

Trigger the [Consolidate](/guides/consolidate) worker -- the background process that compacts recent agent/crew journal activity into semantic rules appended to `.memory/topics/learned-*.md`. Normally runs on a 6h schedule; this command forces an immediate pass.

```bash theme={null}
crewship consolidate <subcommand> [flags]
```

## Subcommands

| Command | Description                           |
| ------- | ------------------------------------- |
| `run`   | Force an immediate consolidation run. |

***

## `crewship consolidate run`

```bash theme={null}
crewship consolidate run
crewship consolidate run --crew cmo2pe4dj0005ba0a129f
crewship consolidate run --since 24h
crewship consolidate run --since 7d
```

| Flag      | Type     | Description                                                                                                                                 |
| --------- | -------- | ------------------------------------------------------------------------------------------------------------------------------------------- |
| `--crew`  | `string` | Limit to a single crew (slug or ID; slugs are resolved automatically). Empty = every crew in the workspace.                                 |
| `--since` | `string` | Only consider journal entries newer than this window. Accepts Go durations (`90m`, `24h`) plus `d`/`w` shorthand (`7d`, `2w`). Default 24h. |

The command takes **no positional args** -- scope is passed via flags. A bare `crewship consolidate run backend-team` fails (`unknown command`).

Sample outputs:

```
Consolidation triggered (worker_id=csd_a1b2c3d4e5f6)
```

```
Accepted, but skipped: no summarizer configured, skipping
```

`Accepted` (202) with a note means the consolidator is wired but lacks a summarizer client -- set `OLLAMA_HOST` and install a chat model.

## Requirements

* **Auth:** `OWNER` or `ADMIN` role (403 otherwise).
* **Ollama:** The consolidator needs a summarizer LLM. In local dev, start Ollama with the external-SSD model path:
  ```bash theme={null}
  OLLAMA_MODELS="/Volumes/SSD 990 PRO/ollama-models" ollama serve
  ```

## Errors

| Exit | Meaning                                                           |
| ---- | ----------------------------------------------------------------- |
| 401  | Not authenticated.                                                |
| 403  | Logged in but not `OWNER` / `ADMIN`.                              |
| 404  | `--crew` points to a crew outside your workspace or soft-deleted. |
| 409  | A consolidation is already running for this workspace.            |
| 503  | Server built without a consolidator instance (dev/test builds).   |

## Journal entries emitted

Even when skipped, the run emits both `system.consolidation_triggered` and `system.consolidation_completed` so the audit trail captures the no-op. See the [Consolidate guide](/guides/consolidate#journal-entries-emitted) for payload shapes.

## Related

* [Consolidate guide](/guides/consolidate).
* [Consolidate API](/api-reference/consolidate).
* [Episodic memory](/guides/episodic-memory) -- indexer runs on the same schedule.