\r \r

Dialogflow CX NLU\r

\r Manage intents and entity types in Google Dialogflow CX via REST API for natural language understanding.\r \r

Prerequisites\r

\r

• Google Cloud project with Dialogflow CX API enabled\r
• Service account or OAuth credentials with Dialogflow API access\r
• gcloud CLI authenticated OR bearer token\r \r

Authentication\r

\r

Option 1: gcloud CLI (recommended)\r

gcloud auth application-default login\r
TOKEN=$(gcloud auth print-access-token)\r
```\r
\r
### Option 2: Service Account\r
```bash\r
export GOOGLE_APPLICATION_CREDENTIALS="/path/to/service-account.json"\r
TOKEN=$(gcloud auth application-default print-access-token)\r
```\r
\r
## API Base URL\r
\r
```\r
https://dialogflow.googleapis.com/v3beta1\r
```\r
\r
Regional endpoints available:\r
- `https://{region}-dialogflow.googleapis.com` (e.g., `us-central1`, `europe-west1`)\r
\r
## Common Operations\r
\r
### List Intents\r
```bash\r
curl -X GET \\r
  "https://dialogflow.googleapis.com/v3beta1/projects/${PROJECT_ID}/locations/${LOCATION}/agents/${AGENT_ID}/intents" \\r
  -H "Authorization: Bearer ${TOKEN}"\r
```\r
\r
### Create Intent\r
```bash\r
curl -X POST \\r
  "https://dialogflow.googleapis.com/v3beta1/projects/${PROJECT_ID}/locations/${LOCATION}/agents/${AGENT_ID}/intents" \\r
  -H "Authorization: Bearer ${TOKEN}" \\r
  -H "Content-Type: application/json" \\r
  -d '{\r
    "displayName": "Book Flight",\r
    "trainingPhrases": [\r
      {\r
        "parts": [{"text": "I want to book a flight"}],\r
        "repeatCount": 1\r
      }\r
    ]\r
  }'\r
```\r
\r
### List Entity Types\r
```bash\r
curl -X GET \\r
  "https://dialogflow.googleapis.com/v3beta1/projects/${PROJECT_ID}/locations/${LOCATION}/agents/${AGENT_ID}/entityTypes" \\r
  -H "Authorization: Bearer ${TOKEN}"\r
```\r
\r
### Create Entity Type\r
```bash\r
curl -X POST \\r
  "https://dialogflow.googleapis.com/v3beta1/projects/${PROJECT_ID}/locations/${LOCATION}/agents/${AGENT_ID}/entityTypes" \\r
  -H "Authorization: Bearer ${TOKEN}" \\r
  -H "Content-Type: application/json" \\r
  -d '{\r
    "displayName": "Cities",\r
    "kind": "KIND_LIST",\r
    "entities": [\r
      {"value": "New York"},\r
      {"value": "Los Angeles"}\r
    ]\r
  }'\r
```\r
\r
## Key Resources\r
\r
| Resource | Description |\r
|----------|-------------|\r
| **Intents** | Classify user utterances and extract parameters |\r
| **Entity Types** | Define structured data extraction patterns |\r
\r
## Quick Reference\r
\r
For detailed API reference:\r
- **Intents**: See [references/intents.md](references/intents.md)\r
- **Entity Types**: See [references/entities.md](references/entities.md)\r
\r
## Scripts\r
\r
- `scripts/nlu.py` — CLI wrapper for intents and entity types operations\r
\r
### Usage\r
```bash\r
python scripts/nlu.py list-intents --agent AGENT_NAME\r
python scripts/nlu.py create-intent --agent AGENT_NAME --intent "Book Flight" --phrases "book a flight,I want to fly"\r
python scripts/nlu.py list-entities --agent AGENT_NAME\r
python scripts/nlu.py create-entity --agent AGENT_NAME --name "Cities" --values "New York,Los Angeles"\r
```\r
\r
## Tips\r
\r
- Use training phrases that cover various ways users might express the intent\r
- Entity types can be system (built-in) or custom\r
- Use KIND_MAP for entities with synonyms, KIND_LIST for simple lists\r