Yard Stats: Semantic Search Tool

The workflow JSON

Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →

{
  "name": "Yard Stats: Semantic Search Tool",
  "nodes": [
    {
      "id": "1",
      "name": "Workflow Input",
      "type": "n8n-nodes-base.executeWorkflowTrigger",
      "typeVersion": 1.1,
      "position": [
        200,
        300
      ],
      "notes": "Called by the 'Semantic Search' tool node inside Yard Stats: Q&A (Call n8n Workflow Tool) -- not meant to be run standalone or scheduled. Expects {query_text, start?, end?, object_types?, limit?} as input -- start/end should be resolved ISO timestamps (the calling Agent already resolved 'last week'/'today' into concrete bounds before calling this tool, same as every other tool in that workflow). Packaged as its own sub-workflow (rather than one HTTP Request Tool node) specifically so the 1024-float embedding vector never has to round-trip through the Agent's own context/tokens -- it's computed and consumed entirely server-side, between these two HTTP calls.",
      "notesInFlow": true,
      "parameters": {
        "workflowInputs": {
          "values": [
            {
              "name": "query_text",
              "type": "string"
            },
            {
              "name": "start",
              "type": "string"
            },
            {
              "name": "end",
              "type": "string"
            },
            {
              "name": "object_types",
              "type": "string"
            },
            {
              "name": "limit",
              "type": "number"
            }
          ]
        }
      }
    },
    {
      "id": "2",
      "name": "Call Embedding Model",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        450,
        300
      ],
      "notes": "Same embedding slot/model (Qwen3-Embedding-0.6B-GGUF, 1024 dims) the AI stage's own embed step uses (ai_worker._embed_text, or n8n's equivalent embed-then-insert step if that flow is active) -- the query text has to be embedded with the same model that wrote the stored sightings' vectors, or cosine distance between them is meaningless.",
      "notesInFlow": true,
      "parameters": {
        "method": "POST",
        "url": "http://REPLACE_WITH_VLM_HOST:REPLACE_WITH_VLM_PORT/REPLACE_WITH_EMBED_SLOT/v1/embeddings",
        "sendBody": true,
        "specifyBody": "json",
        "jsonBody": "={{ { \"input\": $json.query_text } }}",
        "options": {}
      }
    },
    {
      "id": "3",
      "name": "Search Semantic (API)",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        650,
        300
      ],
      "notes": "ingest-worker's POST /search/semantic -- cosine-distance ordered, filtered by the start/end window and object_types the Agent already resolved/chose. Returns the rows as-is; the Agent reads this tool's output directly, no separate parse step needed.",
      "notesInFlow": true,
      "parameters": {
        "method": "POST",
        "url": "http://REPLACE_WITH_INGEST_WORKER_HOST:REPLACE_WITH_INGEST_WORKER_PORT/search/semantic",
        "sendBody": true,
        "specifyBody": "json",
        "jsonBody": "={{ { \"embedding\": $('Call Embedding Model').item.json.data?.[0]?.embedding, \"start\": $('Workflow Input').item.json.start || null, \"end\": $('Workflow Input').item.json.end || null, \"object_types\": $('Workflow Input').item.json.object_types ? $('Workflow Input').item.json.object_types.split(',').map(s => s.trim()) : null, \"limit\": $('Workflow Input').item.json.limit || 10 } }}",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "options": {}
      },
      "credentials": {
        "httpHeaderAuth": {
          "name": "<your credential>"
        }
      }
    }
  ],
  "connections": {
    "Workflow Input": {
      "main": [
        [
          {
            "node": "Call Embedding Model",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Call Embedding Model": {
      "main": [
        [
          {
            "node": "Search Semantic (API)",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "active": false,
  "settings": {
    "executionOrder": "v1"
  }
}

Credentials you'll need

Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.

• httpHeaderAuth

About this workflow

Yard Stats: Semantic Search Tool. Uses executeWorkflowTrigger, httpRequest. Event-driven trigger; 3 nodes.

Source: https://github.com/shuricksumy/frigate-yard-stats/blob/main/n8n/yard-stats-semantic-search-tool.json — original creator credit. Request a take-down →

More Web Scraping workflows → · Browse all categories →

Related workflows

Workflows that share integrations, category, or trigger type with this one. All free to copy and import.