Integrate Captain with Your Data Lake

1
# Step 1: Set your credentials
2
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
3
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
4
5
# Step 2: Create headers dictionary (reuse for all requests)
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
# Step 3: Use headers in every request
12
response = requests.post(
13
    "https://api.runcaptain.com/v1/...",
14
    headers=headers,
15
    data={...}
16
)

1
import requests
2
3
# Set your credentials
4
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
5
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
6
BASE_URL = "https://api.runcaptain.com"
7
8
headers = {
9
    "Authorization": f"Bearer {API_KEY}",
10
    "X-Organization-ID": ORG_ID
11
}
12
13
# 1. Create a database
14
requests.post(
15
    f"{BASE_URL}/v1/create-database",
16
    headers=headers,
17
    data={
18
        'database_name': 'my_documents'
19
    }
20
)
21
22
# 2. Index your S3 bucket (Tagger will scan all files)
23
index_response = requests.post(
24
    f"{BASE_URL}/v1/index-s3",
25
    headers=headers,
26
    data={
27
        'database_name': 'my_documents',
28
        'bucket_name': 'my-s3-bucket',
29
        'aws_access_key_id': 'AKIA...',
30
        'aws_secret_access_key': '...',
31
        'bucket_region': 'us-east-1'
32
    }
33
)
34
35
job_id = index_response.json()['job_id']
36
37
# 3. Monitor indexing progress
38
status = requests.get(
39
    f"{BASE_URL}/v1/indexing-status/{job_id}",
40
    headers=headers
41
)
42
43
# 4. Query your data (DeepQuery executes the search)
44
response = requests.post(
45
    f"{BASE_URL}/v1/query",
46
    headers=headers,
47
    data={
48
        'query': 'What are the main themes in Q4 reports?',
49
        'database_name': 'my_documents'
50
    }
51
)

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/index-s3",
13
    headers=headers,
14
    data={
15
        'database_name': 'my_documents',
16
        'bucket_name': 'my-company-docs',
17
        'aws_access_key_id': 'AKIAIOSFODNN7EXAMPLE',
18
        'aws_secret_access_key': 'wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY',
19
        'bucket_region': 'us-east-1'
20
    }
21
)
22
23
print(response.json())

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/index-s3-file",
13
    headers=headers,
14
    data={
15
        'file_uri': 's3://my-bucket/documents/report_q4_2024.pdf',
16
        'database_name': 'my_documents',
17
        'aws_access_key_id': 'AKIAIOSFODNN7EXAMPLE',
18
        'aws_secret_access_key': 'wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY',
19
        'bucket_region': 'us-east-1'
20
    }
21
)

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
# Load service account JSON
12
with open('path/to/service-account-key.json', 'r') as f:
13
    service_account_json = f.read()
14
15
response = requests.post(
16
    "https://api.runcaptain.com/v1/index-gcs",
17
    headers=headers,
18
    data={
19
        'database_name': 'my_documents',
20
        'bucket_name': 'my-gcs-bucket',
21
        'service_account_json': service_account_json
22
    }
23
)
24
25
print(response.json())

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
with open('service-account-key.json', 'r') as f:
12
    service_account_json = f.read()
13
14
response = requests.post(
15
    "https://api.runcaptain.com/v1/index-gcs-file",
16
    headers=headers,
17
    data={
18
        'file_uri': 'gs://my-bucket/documents/report_q4_2024.pdf',
19
        'database_name': 'my_documents',
20
        'service_account_json': service_account_json
21
    }
22
)

1
import requests
2
3
response = requests.get(
4
    f"https://api.runcaptain.com/v1/indexing-status/{job_id}",
5
    headers={
6
        'Authorization': 'Bearer cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx'
7
    }
8
)
9
10
status = response.json()
11
print(f"Completed: {status['completed']}")
12
print(f"Status: {status['status']}")
13
print(f"Active workers: {status['active_file_processing_workers']}")

1
import requests
2
3
response = requests.get(
4
    f"https://api.runcaptain.com/v1/index-status/{job_id}",
5
    headers={
6
        'Authorization': 'Bearer cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx'
7
    }
8
)
9
10
status = response.json()
11
print(f"Status: {status['status']}")
12
print(f"Progress: {status['progress']['percentage']}%")
13
print(f"ETA: {status['estimated_completion_time']}")

1
import requests
2
3
response = requests.post(
4
    f"https://api.runcaptain.com/v1/index-stop/{job_id}",
5
    headers={
6
        'Authorization': 'Bearer cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx'
7
    }
8
)
9
10
print(response.json())

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/query",
13
    headers=headers,
14
    data={
15
        'query': 'What are the revenue projections for Q4 2024?',
16
        'database_name': 'my_documents',
17
        'include_files': 'true',
18
        'inference': 'true',  # Use Captain's LLM (default)
19
        'topK': '20'  # Limit to top 20 results (optional)
20
    }
21
)
22
23
result = response.json()
24
print(result['response'])

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/query",
13
    headers=headers,
14
    data={
15
        'query': 'What are the revenue projections for Q4 2024?',
16
        'database_name': 'my_documents',
17
        'inference': 'false',  # Return raw chunks instead of AI answer
18
        'topK': '10'  # Get top 10 chunks
19
    }
20
)
21
22
result = response.json()
23
for chunk in result['search_results']:
24
    print(f"{chunk['score']}: {chunk['text'][:100]}...")

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/query",
13
    headers=headers,
14
    data={
15
        'query': 'Summarize all security vulnerabilities found in code reviews',
16
        'database_name': 'my_documents',
17
        'stream': 'true'
18
    },
19
    stream=True
20
)
21
22
# Process streamed chunks
23
for line in response.iter_lines():
24
    if line:
25
        line_text = line.decode('utf-8')
26
        if line_text.startswith('data: '):
27
            print(line_text[6:], end='', flush=True)

1
import requests
2
import uuid
3
4
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
5
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
6
7
idempotency_key = str(uuid.uuid4())
8
9
response = requests.post(
10
    "https://api.runcaptain.com/v1/query",
11
    headers={
12
        'Authorization': f'Bearer {API_KEY}',
13
        'X-Organization-ID': ORG_ID,
14
        'Idempotency-Key': idempotency_key
15
    },
16
    data={
17
        'query': 'What are the main findings?',
18
        'database_name': 'my_documents'
19
    }
20
)
21
22
# Subsequent requests with same key return cached response
23
# No additional processing or token usage

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/create-database",
13
    headers=headers,
14
    data={
15
        'database_name': 'contracts_db'
16
    }
17
)

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/list-databases",
13
    headers=headers
14
)
15
16
databases = response.json()['databases']
17
for db in databases:
18
    print(f"{db['database_name']} - {db['database_id']}")

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/list-files",
13
    headers=headers,
14
    data={
15
        'database_name': 'my_documents',
16
        'limit': 50,
17
        'offset': 0
18
    }
19
)
20
21
files = response.json()['files']
22
for file in files:
23
    print(f"{file['file_name']} - {file['file_type']}")

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/delete-file",
13
    headers=headers,
14
    data={
15
        'database_name': 'my_documents',
16
        'file_id': 'file_abc123'
17
    }
18
)

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/wipe-database",
13
    headers=headers,
14
    data={
15
        'database_name': 'my_documents'
16
    }
17
)
18
19
print(f"Wiped {response.json()['files_deleted']} files")

1
import requests
2
3
API_KEY = "cap_prod_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
4
ORG_ID = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
5
6
headers = {
7
    "Authorization": f"Bearer {API_KEY}",
8
    "X-Organization-ID": ORG_ID
9
}
10
11
response = requests.post(
12
    "https://api.runcaptain.com/v1/delete-database",
13
    headers=headers,
14
    data={
15
        'database_name': 'my_documents'
16
    }
17
)

1
# Good: Separate databases per project
2
databases = [
3
    'project_alpha_docs',
4
    'project_beta_contracts',
5
    'project_gamma_reports'
6
]

1
# Good: Separate databases per department
2
databases = [
3
    'engineering_docs',
4
    'legal_contracts',
5
    'finance_reports'
6
]

1
# Bad: Single database for everything
2
database = 'all_company_documents'  # Hard to manage, slower queries

1
# First index
2
requests.post('.../index-s3', data={'database_name': 'docs', ...})
3
4
# Update some files in S3...
5
6
# Re-index (existing files will be updated)
7
requests.post('.../index-s3', data={'database_name': 'docs', ...})

1
# Good
2
"What are the security vulnerabilities in the authentication module?"
3
4
# Less effective
5
"Tell me about security"

1
# Good
2
"What were the revenue projections mentioned in Q4 2024 reports?"
3
4
# Less effective
5
"What are the numbers?"

1
# Include file metadata to understand sources
2
response = requests.post('.../query', data={
3
    'query': '...',
4
    'include_files': 'true',  # Returns which files were used
5
    ...
6
})

1
# Development
2
API_KEY_DEV = 'cap_dev_...'
3
DATABASE_NAME_DEV = 'docs_dev'
4
5
# Production
6
API_KEY_PROD = 'cap_prod_...'
7
DATABASE_NAME_PROD = 'docs_prod'