search

jonrau1 / ElectricEye

ElectricEye is a multi-cloud, multi-SaaS Python CLI tool for Asset Management, Security Posture Management & Attack Surface Monitoring supporting 100s of services and evaluations to harden your CSP & SaaS environments with controls mapped to over 20 industry, regulatory, and best practice controls frameworks
Apache License 2.0
962 stars 126 forks source link
asset-management attack-surface-management aws aws-audit aws-compliance aws-security cloud-auditing cloud-compliance-reporting cloud-security compliance devsecops gcp-security google-cloud-security multicloud saas-security security-audit security-engineering security-hub security-monitoring security-tools

readme

ElectricEye

ElectricEye is a multi-cloud, multi-SaaS Python CLI tool for Asset Management, Security Posture Management & Attack Surface Monitoring supporting 100s of services and evaluations to harden your CSP & SaaS environments with controls mapped to over 20 industry, regulatory, and best practice controls frameworks.

VulnScan CodeQL EcrBuild OcrBuild DockerHubBuild

Docker Pulls Docker Repo size Issues Contributors License

AWS ECR Gallery Docker Hub

Up here in space
I'm looking down on you
My lasers trace
Everything you do
Judas Priest, 1982

Table of Contents

Workflow

Architecture

Quick Run Down :running: :running:

ElectricEye's core concept is the Auditor which are sets of Python scripts that run Checks per Service dedicated to a specific SaaS vendor or public cloud service provider called an Assessment Target. You can run an entire Assessment Target, a specific Auditor, or a specific Check within an Auditor. After ElectricEye is done with evaluations, it supports over a dozen types of Outputs ranging from an HTML executive report to AWS DocumentDB clusters - you can run multiple Outputs as you see fit.

ElectricEye also uses utilizes other tools such as Shodan.io, Yelp's detect-secrets, VirusTotal, the United States Cyber and Infrastructure Security Agency (CISA) Known Exploited Vulnerability (KEV) Catalog, and NMAP for carrying out its Checks and enriching their findings.

  1. First, clone this repository and install the requirements using pip3: pip3 install -r requirements.txt.

  2. If you are evaluating anything other than your local AWS Account, modify the TOML configuration located in ElectricEye/eeauditor/external_providers.toml, or provide a path to your own with with --toml-path. The TOML file specifies multi-account, mulit-region, credential, and output specifics.

  3. Finally, run the Controller to learn about the various Checks, Auditors, Assessment Targets, and Outputs.

python3 eeauditor/controller.py --help
Usage: controller.py [OPTIONS]

Options:
  -t, --target-provider [AWS|Azure|OCI|GCP|Servicenow|M365|Salesforce|Snowflake]
                                  Public cloud or SaaS assessment target,
                                  ensure that any -a or -c arg maps to your
                                  target provider to avoid any errors. e.g.,
                                  -t AWS -a Amazon_APGIW_Auditor
  -a, --auditor-name TEXT         Specify which Auditor you want to run by
                                  using its name NOT INCLUDING .py. . Use the
                                  --list-checks arg to receive a list.
                                  Defaults to ALL Auditors
  -c, --check-name TEXT           A specific Check in a specific Auditor you
                                  want to run, this correlates to the function
                                  name. Use the --list-checks arg to receive a
                                  list. Defaults to ALL Checks
  -d, --delay INTEGER             Time in seconds to sleep between Auditors
                                  being ran, defaults to 0. Use this argument
                                  to avoid rate limiting
  -o, --outputs TEXT              A list of Outputs (files, APIs, databases,
                                  ChatOps) to send ElectricEye Findings,
                                  specify multiple with additional arguments:
                                  -o csv -o postgresql -o slack  [default:
                                  ocsf_stdout]
  -of, --output-file TEXT         For file outputs such as JSON and CSV, the
                                  name of the file, DO NOT SPECIFY .file_type
                                  [default: output]
  -lo, --list-options             Lists all valid Output options
  -lch, --list-checks             Prints a table of Auditors, Checks, and
                                  Check descriptions to stdout - use this
                                  command for help with populating -a (Auditor
                                  selection) or -c (Check selection) args
  -lco, --list-controls           Lists all ElectricEye controls - that is to
                                  say: the Check Titles - for an Assessment
                                  Target
  -tp, --toml-path TEXT           The full path to the TOML file used for
                                  configure e.g.,
                                  ~/path/to/mydir/external_providers.toml. If
                                  this value is not provided the default path
                                  of ElectricEye/eeauditor/external_providers.
                                  toml is used.
  --help                          Show this message and exit.

For more information see here, you can read the FAQ here, information on Outputs is here or, if you want a more in-depth analysis of the control flow and concepts review the Developer Guide.

Configuring ElectricEye

Refer to sub-headings for per-CSP or per-SaaS setup instructions. Go to Outputs to, well, learn about Outputs and examples.

Public Cloud Service Providers

The following Cloud Service Providers are on the Roadmap

Software-as-a-Service (SaaS) Providers

The following SaaS Providers are on the Roadmap

Cloud Asset Management (CAM)

For more information on ElectricEye's CAM concept of operations and schema, refer to the Asset Management documentation.

Supported Services and Checks

In total there are:

The tables of supported Services and Checks have been migrated to the respective per-Provider setup documentation linked above in Configuring ElectricEye.

ElectricEye on Docker

After configuring ElectricEye for your environment(s) using the TOML configuration, you can instead utilize Docker to run ElectricEye which have images maintained on ECR Public, Oracle Cloud Container Registry (OCR), and Docker Hub. You can read more about the security assurance activities here, in the future more Registries and image signing will be utilized.

Building Images

If you would rather build your own image use the following commands. Be sure to add sudo if you do not have a docker user properly setup in your system.

git clone https://github.com/jonrau1/ElectricEye.git
cd ElectricEye
docker build -t electriceye:local .

From here you can push to your repository of choice, be sure to change the tag from local to whichever tag your repository is expecting or whatever you prefer. Maybe just latest, like a decabillionaire sigma grindset gigachad?

Pulling Images

You can also pull an ElectricEye image from the various repositories, a latest image tag will always be pushed alongside an image tagged with the SHA hash of the workflow ${{ github.sha }} and can be viewed within the various GitHub Action Workflows within the Print Image step.

To pull from the various repositories, use these commands, you can replace latest as you see fit. The dependencies within ElectricEye stay relatively stable until a new cloud or major integration is added. Check the Pull Requests for more information to be sure.

Setting up a Session

NOTE!! You can skip this section if you are using hard-coded credentials in your TOML and if you will not be using any AWS Output or running any AWS Auditors

When interacting with AWS credential stores such as AWS Systems Manager, AWS Secrets Manager and Outputs such as AWS Security and for Role Assumption into the Role specified in the aws_electric_eye_iam_role_name TOML parameter, ElectricEye uses your current (default) Boto3 Session which is derived from your credentials.

Running ElectricEye from AWS Infrastructure that has an attached Role, or running from a location with aws cli credentials already instantiated, this is handled transparently.

When using Docker, you will need to provide Environment Variables directly to the Container.

Ensure that if you will be using AWS SSM (ssm:GetParameter), AWS Secrets Manager (secretsmanager:GetSecretValue), AWS Security Hub (securityhub:BatchImportFindings), Amazon SQS (sqs:SendMessage), and/or Amazon DynamoDB (dynamodb:PutItem) for credentials and Outputs that you have the proper permissions! You will likely also require kms:Decrypt depending if you are using AWS Key Management Service (KMS) Customer-managed Keys (CMKs) for your secrets/parameters encryption.

You will need sts:AssumeRole to assume into the Role specified in the aws_electric_eye_iam_role_name TOML parameter.

You will need to pass in your AWS Region, an AWS Access Key, and an AWS Secret Access Key. If you are NOT using an AWS IAM User with Access Keys you will need to also provide an AWS Session Token which is produced by temporary credentials such as an IAM Role or EC2 Instance Profile.

If you are using a User, proceed to the next step, you will need to have your credentials ready to copy. If you are using an EC2 Instance Profile or an additional IAM Role you will Assume, ensure you have jq installed: apt install -y jq or yum install jq.

  • To Assume an IAM Role and retrieve the temporary credentials
AWS_ACCOUNT_ID=$(aws sts get-caller-identity | jq -r '.Account')
MY_ROLE_NAME='iam-role-name'
TEMP_CREDS=$(aws sts assume-role --role-arn arn:aws:iam::$AWS_ACCOUNT_ID:role/$MY_ROLE_NAME --role-session-name ElectriceyeForDocker)
AWS_ACCESS_KEY=$(echo $TEMP_CREDS | jq -r '.Credentials.AccessKeyId')
AWS_SECRET_KEY=$(echo $TEMP_CREDS | jq -r '.Credentials.SecretAccessKey')
AWS_SESSION_TOKEN=$(echo $TEMP_CREDS | jq -r '.Credentials.SessionToken')
MY_REGION='aws-region-here'
  • To retrieve temporary credentials for an EC2 Instance Profile using Instance Metadata Service Version 1
MY_INSTANCE_PROFILE_ROLE_NAME="my_ec2_role_name"
IMDS_SECURITY_CREDENTIALS=$(curl http://169.254.169.254/latest/meta-data/iam/security-credentials/$MY_INSTANCE_PROFILE_ROLE_NAME)
MY_REGION='my_aws_region'
AWS_ACCESS_KEY=$(echo $IMDS_SECURITY_CREDENTIALS | jq -r '.AccessKeyId')
AWS_SECRET_KEY=$(echo $IMDS_SECURITY_CREDENTIALS | jq -r '.SecretAccessKey')
AWS_SESSION_TOKEN=$(echo $IMDS_SECURITY_CREDENTIALS | jq -r '.Token')
  • To retrieve temporary credentials for an EC2 Instance Profile using Instance Metadata Service Version 2
MY_INSTANCE_PROFILE_ROLE_NAME="my_ec2_role_name"
TOKEN=$(curl -X PUT "http://169.254.169.254/latest/api/token" -H "X-aws-ec2-metadata-token-ttl-seconds: 300")
IMDSV2_SECURITY_CREDENTIALS=$(curl -H "X-aws-ec2-metadata-token: $TOKEN" http://169.254.169.254/latest/meta-data/iam/security-credentials/$MY_INSTANCE_PROFILE_ROLE_NAME)
MY_REGION='my_aws_region'
AWS_ACCESS_KEY=$(echo $IMDSV2_SECURITY_CREDENTIALS | jq -r '.AccessKeyId')
AWS_SECRET_KEY=$(echo $IMDSV2_SECURITY_CREDENTIALS | jq -r '.SecretAccessKey')
AWS_SESSION_TOKEN=$(echo $IMDSV2_SECURITY_CREDENTIALS | jq -r '.Token')

You can also retrieve temporary credentials from Federated identities, read more at the links for AssumeRoleWithWebIdentity or AssumeRoleWithSAML or refer to the larger temporary credential documentation here.

Running ElectricEye Container

Run ElectricEye using the following commands, passing in your Session credentials. Change the commands within the container to evaluate different environments with ElectricEye. Change the value of /path/to/my/external_providers.toml to your exact path, such as ~/electriceye-docker/external_providers.toml for example.

IMPORTANT NOTE If you are using an AWS IAM User with Access Keys, hardcode the values and omit the value for AWS_SESSION_TOKEN!! If you are running this container on an AWS container/Kubernetes service you do not need to provide these values!

sudo docker run \
    --user eeuser:eeuser \
    -e AWS_DEFAULT_REGION=$MY_REGION \
    -e AWS_ACCESS_KEY_ID=$AWS_ACCESS_KEY \
    -e AWS_SECRET_ACCESS_KEY=$AWS_SECRET_KEY \
    -e AWS_SESSION_TOKEN=$AWS_SESSION_TOKEN \
    -v /path/to/my/external_providers.toml:/eeauditor/external_providers.toml \
    electriceye /bin/bash -c "python3 eeauditor/controller.py --help"

To save a local file output such as -o json. -o cam-json, -o csv, or -o html and so on, ensure that you specify a file name that begins with /eeauditor/ as the eeuser within the Docker Image only has permissions within that directory.

To remove the files you cannot use docker cp but you can submit the file to remote APIs you have control of by base64 encoding the output or you can use the Session with AWS S3 permissions to upload the file to S3.

If you are evaluating Oracle Cloud or Google Cloud Platform, your credentials will be locally loaded and you can upload to Oracle Object Storage or Google Cloud Storage buckets, respectively.

BUCKET_NAME="your_s3_bucket_you_have_access_to"
sudo docker run \
    --user eeuser:eeuser \
    -e AWS_DEFAULT_REGION=$MY_REGION \
    -e AWS_ACCESS_KEY_ID=$AWS_ACCESS_KEY \
    -e AWS_SECRET_ACCESS_KEY=$AWS_SECRET_KEY \
    -e AWS_SESSION_TOKEN=$AWS_SESSION_TOKEN \
    -v /path/to/my/external_providers.toml:/eeauditor/external_providers.toml \
    electriceye /bin/bash -c "python3 eeauditor/controller.py -t AWS -o json --output-file /eeauditor/my-aws-findings \
    && aws s3 cp /eeauditor/my-aws-findings.json s3://$BUCKET_NAME/eefindings.json"

For more configuration information ensure you refer back to the per-Provider setup instructions.

Contributing

Refer to the Developer Guide for instructions on how to produce new checks, for new SaaS and CSP support please open an Issue.

Feel free to open PRs and Issues where syntax, grammatic, and implementation errors are encountered in the code base.

ElectricEye is for sale

Hit me up at opensource@electriceye.cloud (I don't actually have a SaaS tool) and I'll gladly sell the rights to this repo and take it down and give you all of the domains and even the AWS Accounts that I use behind the scenes.

Early Contributors

Quick shout-outs to the folks who answered the call early to test out ElectricEye and make it not-a-shit-sandwich.

Alpha Testing:
Beta Testing:

FAQ

This is done Amazonian-style, which is to say, none of these questions are frequently asked and are supposed to help Product Managers figure out if their ideas are good...

That said, some of these questions do get asked. So, you're welcome.

1. What is ElectricEye?

ElectricEye is an agentless Python Command Line Interface (CLI) tool that scans and evaluates Cloud Service Providers (CSPs) and Software-as-a-Service (SaaS) Vendors for service-level configurations. ElectricEye generates a passing or failing finding per resource across multiple checks that align to seucirty posture management best practices as well as other hygeine checks such as resiliency, recovery, performance optimization, and monitoring. ElectricEye covers popular providers such as AWS, GCP, ServiceNow, and more.

2. Who should use ElectricEye?

ElectricEye can be used by any persona within a cloud organization in the security or IT functions such as (but not limited to) Security Engineers, Dev(Sec)Ops Engineers, SREs/Platform Engineers, Architects (various flavors), Governance/Risk/Compliance Analysts, SOC/SecOps Analysts, Cloud Advisors, Offensive Security (Red/Blue/Purple) Teams, and 3rd Party Risk Management Analysts. ElectricEye can also be used by IT Operations, Technology Business Management/ITFM Analysts, and Business Continuity Analysts, and Asset Managers as ElectricEye offers native Cloud Asset Management capabilities.

3. Why should someone use ElectricEye?

ElectricEye should be used by anyone wanting to ensure their cloud vendors and their full breadth of services are configured to ensure the best security hygeine. ElectricEye has the most service coverage offering and is the only dual-use Security Posture Management (SPM) tool that is offered for free for both Cloud Security Posture Management (CSPM) and SaaS Security Posture Management (SSPM). ElectricEye also comes with built-in secrets detection and External Attack Surface Management (EASM) capabilities as well as Cloud Asset Management (CAM) with its own hierarchy to support cross-cloud, cross-boundary asset management and reporting.

4. Is ElectricEye a CSPM?

Yes, ElectricEye is a Cloud Security Posture Management (CSPM) tool, it provides API-based (agentless) scans of cloud infrastructure and ensures services are configured to best practices.

5. Is ElectricEye a SSPM?

Yes, ElectricEye is a SaaS Security Posture Management (SSPM) tool, it provides API-based (agentless) scans of SaaS vendor APIs and ensures users and services are configured to best practices.

6. Is ElectricEye a CIEM?

No, ElectricEye is not a Cloud Infrastructure Entitlement Management (CIEM) tool, while ElectricEye does provide several Identity & Access Management checks and does per-user evaluations for MFA, password rotation, and permissions minimization it is not a CIEM. ElectricEye does not have widespread policy evaluation across multiple identity brokers, providers, and boundaries nor does ElectricEye provide any remediation or Just In Time (JIT) entitlements management capabilities.

7. Is ElectricEye a SIEM?

No, ElectricEye is not a Security Information & Event Management (SIEM) tool. SIEM tools are used to collect, index, and correlate logs, security events, and other semi-structured and structured data for security operations use cases. While ElectricEye findings can be sent to a SIEM, ElectricEye is not a SIEM in its own right.

8. Is ElectricEye an Audit / Compliance Tool?

No, ElectricEye is not directly an Audit or Compliance tool. While every finding is mapped into popular and well-used security compliance regimes such as NIST CSF v1.1 and AICPA 2020 TSCs, ElectricEye only provides best-effort mappings for controls and is not the same as an Auditor or other qualified assessor auditing your environment. ElectricEye can be used as an audit readiness or preparedness tool, you could take samples of findings if the cloud infrastructure controls are important to your overall security program. Controls are technical or administrative (i.e., policy or procedure) countermeasures designed to protect the desired outcomes of a security or privacy program. Controls protect the confidentiality, integrity, and availability of information systems. ElectricEye can help determine if the configurations of your cloud infrastructure meets the "spirit" of the controls but is NOT the same as an attestation, certification, or some other occult ritual.

9. What is "Audit Readiness" ?

ElectricEye uses the term Audit Readiness when communicating the intended use cases for its control frameworks mappings. ElectricEye could be used by qualified assessors to evaluate your environment, it could be used by you to provide as evidence to assessors, but it's best used case is preparing or seeing your readiness for an audit. However, you should already have your own internal controls defined and your own configuration management strategy when it comes to implementing controls. For instance, you may decide it costs too much money and does not offer many security benefits to encrypt all your SQS Queues with AWS KMS CMKs, you have to do that "groundwork" before using ElectricEye to support your internal GRC processes.

10. What control frameworks does ElectricEye support?

The controls frameworks that ElectricEye supports is always being updated as newer versions and mappings are available, as of 21 JUNE 2024 the following standards, frameworks, and legal requirements are supported.

Repository Security

Since ElectricEye is a security tool, it only makes sense to ensure a high-level of security of components are maintained. To that end the following tools are configured for usage. Refer to the build badges and Actions for detailed information about each run.

Syft

See results here!

A CLI tool and Go library for generating a Software Bill of Materials (SBOM) from container images and filesystems.

ElectricEye uses Syft to build an SBOM off of a built ElectricEye Docker Image in CycloneDX format and upload it as an artifact to every succesful GitHub Action run.

Grype

See results here!

A vulnerability scanner for container images and filesystems. Easily install the binary to try it out. Works with Syft, the powerful SBOM (software bill of materials) tool for container images and filesystems.

ElectricEye passes the CycloneDX SBOM from Syft to Grype to perform vulnerability scans on the built Docker image which combines all Python dependencies and built-in methods within the alpine Docker Image parent that ElectricEye uses. Builds with critical vulnerabilities are broken automatically. The results are posted to the GitHub Action and uploaded as sarif to GitHub Security

Dependabot

Dependabot alerts tell you that your code depends on a package that is insecure. If your code depends on a package with a security vulnerability, this can cause a range of problems for your project or the people who use it. You should upgrade to a secure version of the package as soon as possible. If your code uses malware, you need to replace the package with a secure alternative.

ElectricEye uses Dependabot as a Software Composition Analysis (SCA) tool to run daily scans and open Pull Requests in the event that a Docker, Python, or GitHub-Action dependency requires a security patch.

CodeQL

See results here!

CodeQL is the code analysis engine developed by GitHub to automate security checks. You can analyze your code using CodeQL and display the results as code scanning alerts.

ElectricEye uses codeql as a Static Application Security Testing (SAST) tool to scan all Auditors which are written in Python, codeql also looks for secrets in code. It is ran on push and on a schedule.

License

This library is licensed under the Apache-2.0 License. See the LICENSE file.