tracebit-python 0.1.1

CLI for managing Tracebit canary credentials on headless servers

tracebit-python

Python CLI for deploying Tracebit canary credentials on headless servers.

Tracebit provides canary tokens — fake credentials that trigger alerts when used by an attacker. Their official CLI requires browser-based OAuth, which doesn't work on headless servers. This tool uses the Tracebit API directly with pre-generated API tokens.

Installation

pip install tracebit-python

Or from source:

git clone https://github.com/SiteRelEnby/tracebit-python
cd tracebit-python
pip install -e .

Quick Start

1. Get an API token

Log in to community.tracebit.com and create an API token from the web UI.

2. Configure

tracebit configure
# paste your API token when prompted

Or use an environment variable:

export TRACEBIT_API_TOKEN=your-token-here

3. Deploy a canary

tracebit deploy aws --profile staging

This issues canary AWS credentials from Tracebit, writes them to ~/.aws/credentials under the specified profile, and confirms the deployment. If anyone (or anything) uses these credentials, Tracebit fires an alert.

4. Test it

tracebit trigger aws

Runs aws sts get-caller-identity against the canary profile. You should see an alert on the Tracebit dashboard within a few minutes.

5. Keep credentials fresh

Canary credentials expire after ~12 hours. Set up a cron job to refresh them:

# crontab -e
0 */6 * * * /path/to/tracebit refresh --hours 4

Commands

tracebit configure [TOKEN]

Save an API token to ~/.config/tracebit/token. Reads from argument, stdin, or interactive prompt.

tracebit deploy aws

Issue and deploy canary AWS credentials.

Option	Default	Description
--name	hostname	Credential name (shown on Tracebit dashboard)
--profile	staging	AWS profile name in ~/.aws/credentials
--region	from API	AWS region
--labels		Metadata as key=value pairs
--force		Replace existing profile (expires old canary first)

Choose a realistic profile name — staging, backup, legacy-admin, etc. The whole point is for these to look like real credentials to an attacker.

tracebit refresh

Re-issue any credentials expiring within the given threshold. Designed to run from cron.

Option	Default	Description
--hours	2	Refresh credentials expiring within this many hours

tracebit trigger aws

Test-fire a canary by calling aws sts get-caller-identity with the canary profile. Requires the AWS CLI to be installed.

Option	Default	Description
--name	first found	Credential name to trigger

tracebit show

Display deployed canary credentials, their profiles, and expiration status.

tracebit remove

Remove canary credentials locally and expire them on Tracebit's server.

Option	Default	Description
--name	all	Name of credential to remove

Global Options

Option	Description
--token TOKEN	API token (overrides env var and config file)
--base-url URL	Override Tracebit API URL
--json	JSON output (where supported)

Token Resolution

The API token is resolved in this order:

1. --token command-line flag
2. TRACEBIT_API_TOKEN environment variable
3. ~/.config/tracebit/token file

How It Works

1. Issue — requests canary AWS credentials from the Tracebit API
2. Deploy — writes them to ~/.aws/credentials and ~/.aws/config under the chosen profile name
3. Confirm — tells Tracebit the credentials were deployed, so it starts monitoring for usage
4. Alert — any AWS API call using these credentials triggers a detection on the Tracebit dashboard

The credentials have an explicit deny policy — they can't actually do anything in AWS. But any attempt to use them (by an attacker who found them on disk, in a config file, etc.) is logged and alerted on.

File Permissions

• ~/.aws/ directory: 0700
• ~/.aws/credentials, ~/.aws/config: 0600
• ~/.config/tracebit/token: 0600
• ~/.config/tracebit/state.json: 0600

License

Apache License 2.0