iam-ape 1.0.0

IAM AWS Permissions Evaluator

IAM APE

IAM AWS Policy Evaluator

APE takes all of your AWS IAM policies attached to a User, Group, or Role object, and presents you with a single policy, summarizing all of their actual permissions. Taking into account permissions, denials, inherited permissions and permission boundaries!

Setup

Requires Python >= 3.9

From PyPI

1. Run pip install iam-ape
2. Run iam-ape

From source

1. Clone this repository
2. Change directory to iam-ape
3. Run python -m pip install .
4. Run iam-ape

Usage

Prerequisite

Have aws-cli installed on your machine and a profile with aws:GetAccountAuthorizationDetails permissions.
Alternatively, have the json output from aws iam get-account-authorization-details saved to a file.

Before your first run, it's recommended to run iam-ape --update - this updates APE's database with the most current list of all available AWS IAM actions.

The simplest way to use iam-ape is to simply run iam-ape --arn <your-arn-here>
APE will then attempt to fetch the account authorization details, evaluate your permissions, and output a neatly formatted policy to stdout

The --input flag:

If you don't want to fetch the report every time, you can run aws iam get-account-authorization-details by yourself and save the output to a json file. You can then pass that output to APE using the --input flag.

Additional flags:

-o, --output write the output to file instead of stdout
-f, --format (clean|verbose) output the policy in clean, AWS policy-like json format, or a long verbose json containing all specific actions allowed to the entity, the denied actions, and the ineffective (allowed in one place, denied in another) permissions.
-p, --profile the AWS CLI profile to use when fetching Account Authorization Details
-u, --update update APE's database with the most current list of all available AWS IAM actions
-v, --verbose set logging level to DEBUG

Important note: the policy created by this tool might not always be compliant with AWS's constraints. For example, if a user is granted ec2:AttachVolume access to arn:aws:ec2:* by one policy, but denied access to arn:aws:ec2:us-east-1:123456789012:instance/i-123456abc, the resulting policy statement will look like this:

{
    "Action": "ec2:AttachVolume",
    "Resource": "arn:aws:ec2:*",
    "NotResource": "arn:aws:ec2:us-east-1:123456789012:instance/i-123456abc"
}

This statement, having both Resource and NotResource together, is not supported by AWS but makes more sense when trying to understand what the effective permissions of a user are.

Roadmap

• Add an option to supply a resource policy and evaluate whether the entity has access to that resource
• Support additional permissions inherited by Role assumption
• Support SCP Policies