Gable CLI and SDK
gable is Gable on the command line. It publishes contracts, registers data assets and more.
gable --help
Usage: gable [OPTIONS] COMMAND [ARGS]...
Options:
--endpoint TEXT Customer API endpoint for Gable, in the format
https://api.company.gable.ai/
--api-key TEXT API Key for Gable
--version Show the version and exit.
--help Show this message and exit.
Commands:
auth View configured Gable authentication information
contract Validate/publish contracts and check data asset compliance
data-asset Commands for data assets
ping Pings the Gable API to check for connectivity
Getting Started
gable is hosted on PyPi, so to install it just run:
pip install gable
Installing Additional Modules for MySQL and PostgreSQL
Gable's CLI allows you to introspect your database and register tables as data assets within Gable's system. Connecting to these databases require additional packages to communicate with your database(s) of choice.
For MySQL, install the additional packages by running:
pip install 'gable[mysql]'
For PostgreSQL, install the additional packages by running:
pip install 'gable[postgres]'
To install all additional dependencies at once, you can run:
pip install 'gable[all]'
Uploading Lineage Results
Upload a single legacy results file with --results-file:
gable lineage upload --project-root . --namespace dev --results-file results.json
Referenced Prime artifacts require all three explicit file arguments:
gable lineage upload --project-root . --namespace dev \
--strands-file strands.json \
--metadata-file strand-metadata.json \
--code-paths-file code-paths.json
Only explicitly named files are uploaded; sibling files are never discovered.
Incomplete artifact sets, missing files, and conflicting options are rejected before
any upload. Referenced artifacts cannot be combined with --results-file.
The referenced trio is packaged into one gzip-compressed tar stream and sent to
/sca/results through the existing chunk transport. There is no separate
declaration or completion call. The command succeeds only after the server
confirms that all three files are present and valid. Partial transfers are not
usable, and individual files cannot replace an existing set.
Retry the same command after an interruption. The CLI saves a small
<strands-file>.upload-<identity>.json retry-state file before uploading. Keep that
file to resume automatically. State is scoped to file contents and target, so a
different scan or endpoint does not reuse an unrelated run. No scan artifacts are
modified. Use --run-id <UUIDv7> to resume a known run explicitly or to request a
new run; different content under an existing run ID is rejected by the server.
Setting up zsh/bash Autocomplete
The Gable CLI supports shell autocomplete for zsh and bash so you can hit TAB to see available commands and options as you write the command.
To enable it, run the following commands:
_SHELL=zsh # or bash
GABLE_CONFIG_DIR=~/.config/gable
mkdir -p $GABLE_CONFIG_DIR
_GABLE_COMPLETE=${_SHELL}_source gable > $GABLE_CONFIG_DIR/complete.sh
Then add the following to your shell startup scripts (e.g. .zshrc, .bashrc):
source ~/.config/gable/complete.sh
Authentication
To establish an authenticated connection with Gable via the CLI, you need:
- The API endpoint associated with your organization
- An API key that corresponds to the endpoint
In order to find your API key and API endpoint, see the documentation in your Gable web app at (/docs/settings/api_keys).
There are two supported methods for providing this config to the CLI:
Authenticating with CLI Arguments
You have the option to pass the endpoint and API key information directly as arguments during the CLI invocation. For example:
gable --endpoint "https://api.yourorganization.gable.ai" --api-key "yourapikey" ping
Authenticating with Environment Variables
To avoid providing this config every time you execute a command, you can set them as environment variables: GABLE_API_ENDPOINT and GABLE_API_KEY. To make them persistent in your environment, add this to your shell initialization file (e.g. .zshrc or .bashrc):
export GABLE_API_ENDPOINT="https://api.yourorganization.gable.ai"
export GABLE_API_KEY="yourapikey"
Then, you can simply use the CLI as follows:
gable ping
Accessing APIs Behind Proxies (Custom API Headers)
To access the Gable API behind corporate or customer proxies that require custom authentication, users can provide additional HTTP headers using the GABLE_API_HEADERS environment variable. This feature is essential for organizations whose infrastructure enforces proxy authentication or requires custom metadata in API requests.
Usage
Set the GABLE_API_HEADERS environment variable as a JSON string containing your custom headers:
export GABLE_API_HEADERS='{"Authorization": "Bearer YOUR_TOKEN", "X-Proxy-Header": "proxy-value"}'
When set, these headers are automatically included in every API request made by the CLI or client library. Custom headers will override default headers (such as X-API-KEY), allowing flexible integration with proxies, gateways, or custom authentication schemes.
Example
export GABLE_API_KEY=your_api_key
export GABLE_API_ENDPOINT=https://api.example.com
export GABLE_API_HEADERS='{"Authorization": "Bearer YOUR_TOKEN", "X-Proxy-Header": "proxy-value"}'
gable ping
Uploading compliance results
Create a results file with the existing upload envelope:
{"type":"COMPLIANCE","run_id":"01900000-0000-7000-8000-000000000010","data":{}}
gable lineage upload --results-file compliance.json
The run must already have a component and uploaded strands. data is a
provisional object with no domain-specific schema. Results are stored at
strands/<run_id>/compliance.json; another upload replaces the previous document
without creating a run or changing its status.
For large result files, supply the routing metadata explicitly:
gable lineage upload --results-file compliance.json \
--payload-type COMPLIANCE \
--run-id 01900000-0000-7000-8000-000000000010
--payload-type accepts CODE, DATA_STORE, EDGE, CODE_STRANDS,
DATA_COMPONENTS, CODE_DATA_EDGES, COMPLIANCE, or AI_AUGMENTATION; it and
--run-id must be supplied together. In this mode the CLI gzip-compresses and
chunks the file as opaque bytes, adds the metadata to every chunk request header,
and does not parse, validate, or reserialize the JSON. The backend routes using
the headers, then parses the body only when the selected upload handler needs it.
If both options are omitted, the legacy envelope is parsed as before. Explicit
uploads do not require --project-root or --namespace, but their run must
already exist unless the body includes the upload context needed to create it.
1 maintainer
Metadata
1 maintainer
Release files for gable 0.72.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| gable-0.72.0.tar.gz | 218.4 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| gable-0.72.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 498.5 kB
Release files / gable-0.72.0.tar.gz
| Download URL | gable-0.72.0.tar.gz |
|---|---|
| Size | 218.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
1c546356ed07480d6fa1883b76ea8adef8d89b1d8e6d70ce7b281c8c460dd78e
|
|
BLAKE2b-256 checksum How to use checksums |
075284071fe66feff90b63494a9e2f96362834e6dfa99359dbe67f7b6714753c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.21
|
Release files / gable-0.72.0-py3-none-any.whl
| Download URL | gable-0.72.0-py3-none-any.whl |
|---|---|
| Size | 280.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
c14764ab963de3eab8ec7d365f1cb5916a104e5b545e87a9edd5fc7bd9387347
|
|
BLAKE2b-256 checksum How to use checksums |
8d992791dfde837a74e15caa6835bb615691b608f2e64f9294a03a52a06425fa
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.21
|
