Overview
contox_get_context retrieves a single context by its ID, returning all metadata and the full content. Use this when you need to read the actual content of a context that was identified via contox_list_contexts or other tools.
Parameters
| Name | Type | Required | Default | Description |
|---|---|---|---|---|
id | string | Yes | — | The context ID to retrieve |
Return value
Returns the full context object as JSON, including:
| Field | Type | Description |
|---|---|---|
id | string | Unique context ID |
name | string | Display name |
description | string | Brief description |
content | string | Full markdown content of the context |
parentContextId | string or null | Parent context ID |
schemaKey | string or null | Brain schema key |
tier | number or null | Tier level |
state | string or null | State: draft, approved, or deprecated |
tokens | number or null | Approximate token count |
updatedAt | string | ISO timestamp of last update |
Usage examples
Reading a context's content
User: Show me what's in my architecture memory
Claude calls: contox_list_contexts()
// Finds architecture context with id "ctx_def456"
Claude calls: contox_get_context({ id: "ctx_def456" })
Response:
{
"id": "ctx_def456",
"name": "[Memory] Architecture",
"content": "## 2025-04-15\n\n### Next.js 14 App Router Setup\n- TypeScript strict mode...",
...
}
Notes
- Returns the complete content field, which can be large for contexts with extensive history.
- If the context ID does not exist, the tool returns an error.
- Content is in markdown format with date headers separating session entries.