Overview
Fetches complete details for one or more observations by their database IDs. Returns raw tool inputs, outputs, and metadata.Authentication
Requires Bearer token ifauthToken is configured in daemon settings.
Request
Path Parameters
Observation ID(s). Can be a single ID or comma-separated list (max 50)
Response
Array of observation objects
Examples
Single Observation
cURL
Response
Multiple Observations
cURL
Use Cases
Detailed Investigation
After finding relevant observation IDs via GET /search, fetch full details:Context Reconstruction
Reconstruct what happened during a session by fetching all observations in sequence.Implementation Details
Fromdaemon/routes.ts:307-313:
- Accepts comma-separated IDs in the URL path
- Filters invalid IDs (non-numeric or ≤ 0)
- Limits to 50 IDs per request (
MAX_ID_COUNT) - Calls
getFullObservations()which performs a single SQL query - Returns empty array if no valid IDs provided
Error Responses
Invalid IDs
Performance Notes
- Batch retrieval: Use comma-separated IDs instead of multiple requests
- No pagination: This endpoint returns all requested observations at once
- Database indexed: Observation ID is the primary key for fast lookups
Related
- GET /search - Find observation IDs by query
- GET /timeline/:id - Get observations before/after a specific one
- POST /observe - Create new observations