GET /api/recall/unified
Search across both regular memories and procedural runbooks with a single query. Useful for agent runtimes that want one trip per turn instead of two.
curl "https://www.mnueron.com/api/recall/unified?q=deploy%20to%20prod&include=memories,procedurals&limit=20" \
-H "Authorization: Bearer $MNUERON_API_TOKEN"
Query
| Param | Type | Default | Notes |
|---|---|---|---|
q | string | required | Free-text |
namespace | string | — | Filter memories (not procedurals) |
limit | int | 20 | 1..100 |
include | string | memories,procedurals | Comma list — drop one to skip |
Response
{
"query": "deploy to prod",
"memories": [
{ "id": "...", "content": "...", "score": 0.83, ... }
],
"procedurals": [
{
"id": "...",
"title": "Deploy to production",
"summary": "Bump version, push tag, watch GH Actions.",
"trigger_phrases": ["deploy to prod", "ship to production"],
"steps": [...],
"match_kind": "trigger",
"success_count": 12,
"failure_count": 0,
"last_used_at": 1737070000000
}
]
}
match_kind is either trigger (the query matched a trigger_phrases[] exact entry) or title-fuzzy.
Memory hits are auto-logged as source: "unified-recall" recall events — you don't need to call /api/recall separately.