### Install Development Dependencies

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Install dependencies required for development, typically listed in requirements.txt.

```bash
pip install -r requirements.txt
```

--------------------------------

### Install Dependencies with Pip

Source: https://github.com/omnifactai/omnifact-cli/blob/master/CLAUDE.md

Install the package in editable mode and development dependencies. Ensure you are in a virtual environment.

```bash
pip install -e .

```

```bash
pip install -r requirements.txt

```

--------------------------------

### Install Omnifact CLI Package

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Install the Omnifact CLI package in editable mode. This allows for development and testing.

```bash
pip install -e .
```

--------------------------------

### Configuration Commands

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Commands for setting and getting the API key and Connect URL. These can be persisted in a config file or overridden via environment variables.

```APIDOC
## Configuration Commands

### `config set-api-key` / `config get-api-key`

Persists or retrieves the Omnifact API key in `~/.omnifact_cli.ini`. The environment variable `OMNIFACT_API_KEY` takes precedence.

```bash
# Store API key permanently
omnifact-cli config set-api-key sk-abc123

# Verify stored key
omnifact-cli config get-api-key
# Output: Current API key: sk-abc123
```

### `config set-connect-url` / `config get-connect-url`

Overrides the default Omnifact Connect API base URL (`https://connect.omnifact.ai`). The environment variable `OMNIFACT_CONNECT_URL` also overrides the stored value.

```bash
# Point to a custom Connect instance
omnifact-cli config set-connect-url https://connect.my-company.com

# Verify
omnifact-cli config get-connect-url
# Output: Current Connect URL: https://connect.my-company.com
```
```

--------------------------------

### Clone Omnifact CLI Repository

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Clone the Omnifact CLI repository to your local machine. This is the first step for installation.

```bash
git clone https://github.com/OmnifactAI/omnifact-cli.git
cd omnifact-cli
```

--------------------------------

### List Supported File Types

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Query the Omnifact API to get a list of all supported MIME types and their corresponding file extensions for document uploads.

```bash
omnifact-cli list-supported-types

# Output:
# Supported file types:
#   application/pdf: pdf
#   text/plain: txt, text
#   application/vnd.openxmlformats-officedocument.wordprocessingml.document: docx
#   text/markdown: md
```

--------------------------------

### Start an interactive chat session

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Initiate a persistent multi-turn conversation with an AI endpoint using the `chat -i` command. Conversation history is maintained in memory. Type 'exit' or 'quit' to end.

```bash
omnifact-cli chat --endpoint-id endpoint-abc -i

# Output:
# Interactive chat started. Type 'exit' or 'quit' to end the session.
# --------------------------------------------------
# You: What is the Omnifact API?
# Assistant: The Omnifact API allows you to manage documents and interact
# with AI assistants programmatically...
# You: Can you give me a Python example?
# Assistant: Sure! Here is how you can list documents using the requests library...
# You: exit
# Goodbye!
```

--------------------------------

### Get Document Details

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Retrieve details for a specific document using its ID.

```bash
omnifact-cli get-document DOCUMENT_ID
```

--------------------------------

### Get supported file types using OmnifactAPI

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Retrieve a list of file types supported by Omnifact, including their MIME types and associated extensions, using the `get_supported_file_types` method.

```python
types = api.get_supported_file_types()
for t in types:
    print(t["mimeType"], t["extensions"])
```

--------------------------------

### Chat (Interactive Mode)

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Starts a persistent multi-turn conversation session with an AI endpoint. Conversation history is maintained in memory for the duration of the session, enabling follow-up questions. Type `exit` or `quit` to end the session.

```APIDOC
## chat (interactive mode)

### Description
Starts a persistent multi-turn conversation session with an AI endpoint. Conversation history is maintained in memory for the duration of the session, enabling follow-up questions. Type `exit` or `quit` to end the session.

### Usage
```bash
omnifact-cli chat --endpoint-id <endpoint_id> -i
```

### Example
```bash
omnifact-cli chat --endpoint-id endpoint-abc -i

# Output:
# Interactive chat started. Type 'exit' or 'quit' to end the session.
# --------------------------------------------------
# You: What is the Omnifact API?
# Assistant: The Omnifact API allows you to manage documents and interact
# with AI assistants programmatically...
# You: Can you give me a Python example?
# Assistant: Sure! Here is how you can list documents using the requests library...
# You: exit
# Goodbye!
```
```

--------------------------------

### Display Help Information

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Display all available commands and their usage information.

```bash
omnifact-cli --help
```

--------------------------------

### Create and Activate Virtual Environment

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Create a Python virtual environment and activate it. This isolates project dependencies.

```bash
python3 -m venv venv
source venv/bin/activate # On Windows, use venv\Scripts\activate
```

--------------------------------

### Use Environment Variables for Configuration

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Configure API key and Connect URL using environment variables. Useful for scripts and CI/CD.

```bash
OMNIFACT_API_KEY=your_api_key OMNIFACT_CONNECT_URL=https://your-connect-url.com omnifact-cli list-documents --space-id YOUR_SPACE_ID
```

--------------------------------

### Initialize OmnifactAPI and list documents

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Instantiate the OmnifactAPI class with your API key and base URL. Use the `get_documents` method to retrieve documents from a space, handling pagination with offset and limit.

```python
from omnifact_cli.api import OmnifactAPI

api = OmnifactAPI(api_key="sk-abc123", base_url="https://connect.omnifact.ai")

# List documents (paginated)
page = api.get_documents("space-xyz-789", offset=0, limit=100)
for doc in page["items"]:
    print(doc["id"], doc["name"])
```

--------------------------------

### List Supported File Types

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

List all file types that the Omnifact CLI supports for upload.

```bash
omnifact-cli list-supported-types
```

--------------------------------

### Upload Document with Options

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Upload a file to a space, with options for a custom display name, JSON metadata, and explicit file encoding. The encoding is sent as a `Content-Transfer-Encoding` header.

```bash
# Basic upload
omnifact-cli upload-document --space-id space-xyz-789 --file ./report.pdf

# Upload with custom name and metadata
omnifact-cli upload-document \
  --space-id space-xyz-789 \
  --file ./report.pdf \
  --name "Q4 Financial Report" \
  --metadata '{"url": "https://example.com/reports/q4.pdf", "author": "finance-team"}'

# Upload a legacy Windows-encoded text file
omnifact-cli upload-document \
  --space-id space-xyz-789 \
  --file ./legacy-export.txt \
  --encoding windows-1252

# Output:
# Document uploaded successfully. ID: doc-004
```

--------------------------------

### Configure Omnifact API Key

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Set your Omnifact API key using the CLI. This is required for authentication.

```bash
omnifact-cli config set-api-key YOUR_API_KEY
```

--------------------------------

### Configure API Key with Omnifact CLI

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Store or retrieve your Omnifact API key using the `config set-api-key` and `config get-api-key` commands. The `OMNIFACT_API_KEY` environment variable takes precedence.

```bash
omnifact-cli config set-api-key sk-abc123

omnifact-cli config get-api-key
# Output: Current API key: sk-abc123

export OMNIFACT_API_KEY=sk-abc123
omnifact-cli list-documents --space-id my-space

OMNIFACT_API_KEY=sk-abc123 omnifact-cli list-documents --space-id my-space
```

--------------------------------

### Configure Omnifact Connect URL

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Optionally, set your Omnifact Connect URL using the CLI. This specifies the API endpoint.

```bash
omnifact-cli config set-connect-url YOUR_CONNECT_URL
```

--------------------------------

### Run Tests

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Execute the project's test suite using pytest.

```bash
pytest
```

--------------------------------

### Upload a Document

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Upload a document to a specified space. Supports custom names and metadata.

```bash
omnifact-cli upload-document --space-id YOUR_SPACE_ID --file /path/to/your/document.pdf
```

```bash
omnifact-cli upload-document --space-id YOUR_SPACE_ID --file /path/to/your/document.pdf --name "Document Name" --metadata '{"url": "https://example.com/docs/document.pdf"}'
```

```bash
omnifact-cli upload-document --space-id YOUR_SPACE_ID --file /path/to/your/document.txt --encoding windows-1252
```

--------------------------------

### Upload a document from disk using OmnifactAPI

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Use the `upload_document` method to upload a file from disk. Specify the space ID, file path, document name, and optional metadata. Encoding is auto-detected by default.

```python
result = api.upload_document(
    space_id="space-xyz-789",
    file_path="./report.pdf",
    name="Q4 Report",
    metadata={"source": "finance"},
    encoding=None  # auto-detect
)
print("Uploaded:", result["id"])
```

--------------------------------

### List Documents in a Space

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Retrieve and display all documents within a specified space. The command automatically handles pagination for large numbers of documents.

```bash
omnifact-cli list-documents --space-id space-xyz-789

# Output:
# ID: doc-001, Name: architecture-overview.pdf
# ID: doc-002, Name: api-reference.md
# ID: doc-003, Name: release-notes.txt
```

--------------------------------

### Configure Omnifact Connect URL with CLI

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Override the default Omnifact Connect API base URL using `config set-connect-url` and `config get-connect-url`. The `OMNIFACT_CONNECT_URL` environment variable also provides an override.

```bash
omnifact-cli config set-connect-url https://connect.my-company.com

omnifact-cli config get-connect-url
# Output: Current Connect URL: https://connect.my-company.com

OMNIFACT_CONNECT_URL=https://staging.omnifact.ai omnifact-cli list-documents --space-id my-space
```

--------------------------------

### Chat with AI Assistant (Single Question)

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Ask a single question to an AI assistant. Requires the endpoint ID.

```bash
omnifact-cli chat --endpoint-id YOUR_ENDPOINT_ID "What is clean code?"
```

--------------------------------

### Chat with AI Assistant (Interactive Mode)

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Engage in an interactive chat session with an AI assistant. Type 'exit' or 'quit' to end.

```bash
omnifact-cli chat --endpoint-id YOUR_ENDPOINT_ID -i
```

--------------------------------

### Format Code with Black and Isort

Source: https://github.com/omnifactai/omnifact-cli/blob/master/CLAUDE.md

Format Python code using Black for style and Isort for import sorting.

```bash
black omnifact_cli tests

```

```bash
isort omnifact_cli tests

```

--------------------------------

### List Documents in a Space

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

List all documents within a specified space. Requires the space ID.

```bash
omnifact-cli list-documents --space-id YOUR_SPACE_ID
```

--------------------------------

### Document Management Commands

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Commands for managing documents within Omnifact spaces, including listing, uploading, retrieving, renaming, and deleting.

```APIDOC
## Document Management

### `list-documents`

Retrieves all documents in a given space, automatically paginating through results in batches of 100. Prints each document's ID and name to stdout.

```bash
omnifact-cli list-documents --space-id space-xyz-789

# Output:
# ID: doc-001, Name: architecture-overview.pdf
# ID: doc-002, Name: api-reference.md
# ID: doc-003, Name: release-notes.txt
```

### `upload-document`

Uploads a file to a space. Supports an optional display name, JSON metadata, and explicit file encoding. Returns the new document's ID on success.

```bash
# Basic upload
omnifact-cli upload-document --space-id space-xyz-789 --file ./report.pdf

# Upload with custom name and metadata
omnifact-cli upload-document \
  --space-id space-xyz-789 \
  --file ./report.pdf \
  --name "Q4 Financial Report" \
  --metadata '{"url": "https://example.com/reports/q4.pdf", "author": "finance-team"}'

# Upload a legacy Windows-encoded text file
omnifact-cli upload-document \
  --space-id space-xyz-789 \
  --file ./legacy-export.txt \
  --encoding windows-1252

# Output:
# Document uploaded successfully. ID: doc-004
```

### `get-document`

Fetches and displays the details of a single document by its ID, including processing status and any attached metadata.

```bash
omnifact-cli get-document doc-001

# Output:
# ID: doc-001
# Name: architecture-overview.pdf
# Status: ready
# Metadata: {
#   "url": "https://example.com/docs/architecture.pdf"
# }
```

### `rename-document`

Updates the display name of an existing document. Requires the document ID as a positional argument and the new name via `--name`.

```bash
omnifact-cli rename-document doc-001 --name "System Architecture v2.pdf"

# Output:
# Document renamed successfully to: System Architecture v2.pdf
```

### `delete-document`

Permanently removes a single document by ID. No confirmation is requested.

```bash
omnifact-cli delete-document doc-003

# Output:
# Document doc-003 deleted successfully.
```

### `list-supported-types`

Queries the API for all currently supported MIME types and their corresponding file extensions for document upload.

```bash
omnifact-cli list-supported-types

# Output:
# Supported file types:
#   application/pdf: pdf
#   text/plain: txt, text
#   application/vnd.openxmlformats-officedocument.wordprocessingml.document: docx
#   text/markdown: md
```
```

--------------------------------

### Run Pytest Commands

Source: https://github.com/omnifactai/omnifact-cli/blob/master/CLAUDE.md

Execute all tests, a single test file, a specific test, or tests with coverage reporting using pytest.

```bash
pytest

```

```bash
pytest tests/test_cli.py

```

```bash
pytest tests/test_cli.py::test_list_documents

```

```bash
pytest --cov=omnifact_cli

```

--------------------------------

### Upload a document from memory using OmnifactAPI

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Use `upload_document_from_memory` to upload content directly from memory, such as generated text. Provide file data, file name, content type, and document name.

```python
import io
content = b"This is an in-memory document."
result = api.upload_document_from_memory(
    space_id="space-xyz-789",
    file_data=io.BytesIO(content),
    file_name="generated.txt",
    content_type="text/plain",
    name="Generated Doc"
)
print("Uploaded from memory:", result["id"])
```

--------------------------------

### Send a single message to an AI endpoint

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Use the `chat` command in single-question mode to send a message to an AI endpoint and stream the response. The output is displayed token-by-token as it arrives.

```bash
omnifact-cli chat --endpoint-id endpoint-abc "What is clean code?"

# Output (streamed):
# Clean code is code that is easy to read, understand, and maintain.
# It follows consistent conventions, uses meaningful names, and expresses
# intent clearly without unnecessary complexity...
```

--------------------------------

### Chat (Single-Question Mode)

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Sends a single message to an Omnifact AI endpoint and streams the response to stdout using Server-Sent Events. The assistant's reply is printed token-by-token as it arrives.

```APIDOC
## chat (single-question mode)

### Description
Sends a single message to an Omnifact AI endpoint and streams the response to stdout using Server-Sent Events. The assistant's reply is printed token-by-token as it arrives.

### Usage
```bash
omnifact-cli chat --endpoint-id <endpoint_id> "<your_question>"
```

### Example
```bash
omnifact-cli chat --endpoint-id endpoint-abc "What is clean code?"

# Output (streamed):
# Clean code is code that is easy to read, understand, and maintain.
# It follows consistent conventions, uses meaningful names, and expresses
# intent clearly without unnecessary complexity...
```
```

--------------------------------

### Rename a Document

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Rename an existing document by providing its ID and the new name.

```bash
omnifact-cli rename-document DOCUMENT_ID --name "New Document Name.pdf"
```

--------------------------------

### Perform streaming chat with OmnifactAPI

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Initiate a streaming chat with an AI endpoint using `api.chat` with `stream=True`. Iterate over the response lines and decode them to display the streamed output.

```python
response = api.chat("endpoint-abc", "Summarize clean code principles", stream=True)
for line in response.iter_lines():
    if line:
        print(line.decode("utf-8"))
```

--------------------------------

### Rename Document

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Update the display name of an existing document. Requires the document ID and the new name specified with the `--name` flag.

```bash
omnifact-cli rename-document doc-001 --name "System Architecture v2.pdf"

# Output:
# Document renamed successfully to: System Architecture v2.pdf
```

--------------------------------

### Purge All Documents from a Space

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Remove all documents from a specified space. Requires confirmation.

```bash
omnifact-cli purge --space-id YOUR_SPACE_ID
```

--------------------------------

### OmnifactAPI Python Class

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

The OmnifactAPI class provides programmatic access to Omnifact functionalities, including document management and AI chat, via a Python library.

```APIDOC
## OmnifactAPI Python Class

### Description
The `OmnifactAPI` class in `omnifact_cli/api.py` can also be used directly as a Python library for programmatic integration. It wraps all HTTP communication in a `requests.Session` and supports both streaming and non-streaming chat responses.

### Initialization
```python
from omnifact_cli.api import OmnifactAPI

api = OmnifactAPI(api_key="<your_api_key>", base_url="<your_base_url>")
```

### Document Operations

#### List Documents (Paginated)
```python
page = api.get_documents("<space_id>", offset=0, limit=100)
for doc in page["items"]:
    print(doc["id"], doc["name"])
```

#### Upload Document from Disk
```python
result = api.upload_document(
    space_id="<space_id>",
    file_path="./<your_file.pdf>",
    name="<document_name>",
    metadata={"source": "<source_name>"},
    encoding=None  # auto-detect
)
print("Uploaded:", result["id"])
```

#### Upload Document from Memory
```python
import io
content = b"This is an in-memory document."
result = api.upload_document_from_memory(
    space_id="<space_id>",
    file_data=io.BytesIO(content),
    file_name="<generated_file.txt>",
    content_type="text/plain",
    name="<Generated Doc>"
)
print("Uploaded from memory:", result["id"])
```

#### Rename Document
```python
updated = api.update_document("<document_id>", "<New Name.pdf>")
print("Renamed to:", updated["name"])
```

#### Delete Document
```python
api.delete_document("<document_id>")
```

#### Get Supported File Types
```python
types = api.get_supported_file_types()
for t in types:
    print(t["mimeType"], t["extensions"])
```

### AI Chat Operations

#### Streaming Chat
```python
response = api.chat("<endpoint_id>", "Summarize clean code principles", stream=True)
for line in response.iter_lines():
    if line:
        print(line.decode("utf-8"))
```

#### Non-Streaming Chat
```python
result = api.chat("<endpoint_id>", "What is Omnifact?", stream=False)
print(result)
```
```

--------------------------------

### Rename a document using OmnifactAPI

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Update a document's name using the `update_document` method, providing the document ID and the new name.

```python
updated = api.update_document("doc-001", "New Name.pdf")
print("Renamed to:", updated["name"])
```

--------------------------------

### Delete a Document

Source: https://github.com/omnifactai/omnifact-cli/blob/master/README.md

Delete a document from a space using its ID.

```bash
omnifact-cli delete-document DOCUMENT_ID
```

--------------------------------

### Delete a document using OmnifactAPI

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Remove a document from a space by its ID using the `delete_document` method.

```python
api.delete_document("doc-003")
```

--------------------------------

### Purge Documents

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Deletes all documents in a specified space after listing them and prompting for confirmation. Handles pagination automatically for large spaces.

```APIDOC
## purge

### Description
Deletes all documents in a space after listing them and prompting for confirmation. Pagination is handled automatically, so large spaces are fully cleared regardless of document count.

### Usage
```bash
omnifact-cli purge --space-id <space_id>
```

### Example
```bash
omnifact-cli purge --space-id space-xyz-789

# Output:
# The following documents will be deleted:
# ID: doc-001, Name: architecture-overview.pdf
# ID: doc-002, Name: api-reference.md
# Are you sure you want to delete all these documents? [y/N]: y
# Deleted document: doc-001
# Deleted document: doc-002
# All documents have been purged from the space.

# Cancel the operation
omnifact-cli purge --space-id space-xyz-789
# Are you sure you want to delete all these documents? [y/N]: n
# Operation cancelled.
```
```

--------------------------------

### Perform non-streaming chat with OmnifactAPI

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Send a chat message to an AI endpoint without streaming using `api.chat` with `stream=False`. The complete response is returned as a single result.

```python
result = api.chat("endpoint-abc", "What is Omnifact?", stream=False)
print(result)
```

--------------------------------

### Delete Document

Source: https://context7.com/omnifactai/omnifact-cli/llms.txt

Permanently remove a single document by its ID. This action is irreversible and does not prompt for confirmation.

```bash
omnifact-cli delete-document doc-003

# Output:
# Document doc-003 deleted successfully.
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.