Skip to main content

Overview

Retrieve the complete, untruncated content of a specific observation. Use this tool after mem_search or mem_timeline when you need to see the full details of a memory entry.
This tool is part of the agent profile and uses deferred loading to optimize performance.

Parameters

id
number
required
The observation ID to retrieve (from mem_search or mem_timeline results)

Response

result
string
Formatted observation with complete content and metadata
The response includes:
  • Observation ID, type, and title
  • Full content (untruncated)
  • Session ID and project
  • Scope (project or personal)
  • Topic key (if set)
  • Tool name (if captured automatically)
  • Duplicate count and revision count
  • Last seen timestamp
  • Created timestamp

Usage Example

{
  "id": 42
}
Response:
#42 [bugfix] Fixed FTS5 syntax error on special chars
**What**: Wrapped each search term in quotes before passing to FTS5 MATCH
**Why**: Users typing queries like 'fix auth bug' would crash because FTS5 interprets special chars as operators
**Where**: internal/store/store.go — sanitizeFTS() function
**Learned**: FTS5 MATCH syntax is NOT the same as LIKE — always sanitize user input
Session: session-abc-123
Project: engram
Scope: project
Topic: bug/fts5-sanitization
Duplicates: 0
Revisions: 2
Created: 2026-03-01T14:23:45Z

When to Use

  • After search: When mem_search returns truncated results and you need the full content
  • Timeline drilling: After mem_timeline to see complete details of a specific entry
  • Content review: When you need to read the full observation including all fields

Progressive Disclosure Pattern

  1. Start with mem_search to find relevant memories (returns truncated content)
  2. Review search results to identify the observation you need
  3. Use mem_get_observation to retrieve the full content
  4. Optionally use mem_timeline to see chronological context around that observation
The observation ID is shown in square brackets in mem_search and mem_timeline results: [42]
  • mem_search - Search for observations (returns truncated content)
  • mem_timeline - View chronological context around an observation
  • mem_update - Update an observation after retrieving it

Build docs developers (and LLMs) love