kairo-engine 1.0.1

A configuration-driven engine that dynamically mutates incoming payloads based on conditional rules defined in YAML or JSON.

kairo-engine

Configuration-driven engine that dynamically mutates incoming payloads based on conditional rules defined in YAML or JSON.

---

--Overview--

Kairo is a configuration engine that modifies payload requests into a new one. With Canaries defined in a YAML-config or JSON-config, you can specify a matching Condition that applies Mutations if the condition is met i.e - if an path with a value is found. A mutation then applies any mutation given it finds a path and mutates it to a jmes-path search condition. Refer to 'How to Run' and 'Example Usage' for further examples.

---

--How To Run--

To run the engine, you need to create a Kairo() instance, load it properly with a yaml or json config - then run any condition or mutation(s) into incoming requests.

E.g -

kairo = Kairo()
kairo.load_from_yaml("canaries.yaml")
result = kairo.run({"some": "request"})

You can also create your own Condition / Mutations objects dynamically and run it through the Canary evaluator. To do so you first define a Match, and Filter objects to create your condition/mutation instances, initialize a canary object and evaluate it with an incoming payload request as such:

1️⃣ Get your request

original_request = {"foo": {"bar": 1, "baz": 2}}

2️⃣ Create your Match and Filter objects

match_obj = Match(path="foo.bar", with_value=1) # create match condition
filter_obj = Filter(path="foo.bar", to_value="foo.baz") # create filter condition

3️⃣ Create your Condition, Mutation, and (optional) added config

condition = Condition(match=match_obj)
mutation = Mutation(filter=filter_obj)
config = {"add": "me"}

4️⃣ Create Canary object and run evaluation

canary = Canary("test_foo_bar_baz", condition=condition, mutations=[mutation], config=config)
new_request = canary.evaluate(original_request, override=False)

5️⃣ Result

new_request = {
    "foo": {
        "bar": 2,
        "baz": 2
    },
    "add": "me"
}

--- To run it through a YAML-config or JSON-config the format should be as follows:

YAML Config:

global_config:
  name: (str)

  canaries:
    - name: (str)
      condition:
        match:
          path: (str)
          with_value: (Any)
      mutations:
        - filter:
            path: (str)
            to_value: (str)
        ...
    ...

      config:
          ...

JSON Config:

{
    "global_config": {
        "name": (str),
        "canaries": [
            {
                "name": (str),
                "condition": {
                    "match": {
                        "path": (str),
                        "with_value": (Any)
                    }
                },
                "mutations": [
                    {
                        "filter": {
                            "path": (str),
                            "to_value": (str)
                        }
                    } ...
                ]
            } ...
        ],
        "config": (Any)
    }
}

---

Testing

This is a dockerized python-3.12 and can be ran with Makefile commands as follows:

1. "make ps" -> docker-compose ps

2. "make build" -> docker-compose build

3. "make build-clean" -> docker-compose build-clean

4. "make exec" -> docker-compose exec server bash

5. "make up" -> docker-compose up

6. "make up-test" -> docker-compose up test

7. "make down" -> docker-compose down

Testing can be done inside of the container. Once it's spun up you can exec into it and run tests there.

1) Make a build or new clean build.
2) Exec into the container.
3) Inside the container run commands: `python3.12 -m pytest tests`

---

License

Released under the MIT License — see LICENSE.