We should have a way to have ctfcli verify that a given installed challenge matches the local state. This could help in situations where perhaps an admin deletes files or changes a description.

When creating a new challenge (blank) the state of the challenge is by default set to "Hidden".

Upon adding and installing the challenge and opening CTFd the challenge will be "Visible" and not "Hidden" as per the challenge.yml.
When changing the state in the challenge.yml to "Visible" (and syncing the change) the challenge will be set to "Hidden" in CTFd.

ctf challenge deploy chal registry://myregistry fails with TypeError: cannot unpack non-iterable NoneType object, because it tries to unpack data that isn't returned when deploying to a registry:

I've seen people include files in dist and then either not include them in challenge.yml or write incorrect keys in challenge.yml that can't be found.

For example:

dist:
  - dist/source.zip

We should have a command or switches that will show what we are reading/loading.

Some useful information:

• Target URL
• API Key
• Location of the .ctf/config file
• Contents of the .ctf/config file

We can include a healthcheck field in challenge.yml that specifies a script for us to execute to check if a challenge is viable. We can pass it the connection info in CTFd as well.

Add the ability to specify 'Case Insensitive' flags in the spec.

Hi,

I may be wrong but it's not possible to add a git repo with ctf challenge add {repo_url} at first. This command relies on git subtree which is a shell script. this script calls ensure_clean to check the HEAD reference but our repo is empty at this point.

I would add a git commit --allow-empty -m "Project created" after the git init in main.py.

git subtree was something being research when ctfcli was being implemented but it was discounted for one reason or another. I think it's worth revisiting that discussion as well as seeing how we can vary challenge repos by branch. We would need to record branch information in order to support git subtrees anyway.

We should have a ctf challenge test command. Some ideas:

• Test that Dockerfiles build
• Test the created image
• Test chal is solveable

Add a command to install and Sync commands that allow you install all challenges in a specified directory. Allowing it to loop through the directories recursively would be nice too. That way I can store my challenges in different directories based on category.

IE: I would run ctf challenge install Challenges

where the challenges folder looks like this.
Challenges
Crypto
rsa1
challenge.yml
rsa2
challenge.yml
Binary
intoverflow
challenge.yml
bufferoverflow
challenge.yml
...

It's unclear how to remove a challenge. Especially if it's deleted from the local event repo.

ctfcli is installed with pip and initialized. Then a new challenge is created with ctf challenge new web and given the name "Hello". On ctf challenge install Hello the following result appears:

Found Hello/challenge.yml
Loaded Hello
Traceback (most recent call last):
  File "/usr/local/bin/ctf", line 8, in <module>
    sys.exit(main())
  File "/usr/local/lib/python3.8/site-packages/ctfcli/__main__.py", line 74, in main
    fire.Fire(CTFCLI)
  File "/usr/local/lib/python3.8/site-packages/fire/core.py", line 138, in Fire
    component_trace = _Fire(component, args, parsed_flag_args, context, name)
  File "/usr/local/lib/python3.8/site-packages/fire/core.py", line 466, in _Fire
    component, remaining_args = _CallAndUpdateTrace(
  File "/usr/local/lib/python3.8/site-packages/fire/core.py", line 675, in _CallAndUpdateTrace
    component = fn(*varargs, **kwargs)
  File "/usr/local/lib/python3.8/site-packages/ctfcli/cli/challenges.py", line 94, in install
    installed_challenges = load_installed_challenges()
  File "/usr/local/lib/python3.8/site-packages/ctfcli/utils/challenge.py", line 28, in load_installed_challenges
    return s.get("/api/v1/challenges?view=admin", json=True).json()["data"]
  File "/usr/local/lib/python3.8/site-packages/requests/models.py", line 897, in json
    return complexjson.loads(self.text, **kwargs)
  File "/usr/local/Cellar/[email protected]/3.8.5/Frameworks/Python.framework/Versions/3.8/lib/python3.8/json/__init__.py", line 357, in loads
    return _default_decoder.decode(s)
  File "/usr/local/Cellar/[email protected]/3.8.5/Frameworks/Python.framework/Versions/3.8/lib/python3.8/json/decoder.py", line 337, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/local/Cellar/[email protected]/3.8.5/Frameworks/Python.framework/Versions/3.8/lib/python3.8/json/decoder.py", line 355, in raw_decode
    raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

What are the current plans for the image field in the chal spec? One thing to keep in mind for deployment is that some chals might want multiple containers e.g. a sql db + webserver, so perhaps specifying a single image is not enough. For the CTF we recently ran, we had a docker-compose file for each challenge.

The function install() in ctfcli/cli/challenge.py implements a check to see if there is already a challenge with the same name. This check does this as follows:

if c["name"] == challenge["name"]:
    click.secho(
        "Already found existing challenge with same name. Perhaps you meant sync instead of install?",
        fg="red",
    )

As a result, ctfcli refuses to install challenges with the same name regardless of the category.

I have been using a modified variant of ctfcli that determines if a challenge is already present based on the challenge name and challenge category.

if c["name"] == challenge["name"] and c["category"] == challenge["category"]:
    click.secho(
        "Already found existing challenge with same name. Perhaps you meant sync instead of install?",
        fg="red",
    )

The same check is also present in the function sync_challenge() in ctfcli/utils/challenge.py and should be changed in similar fashion to sync the correct challenge.

I guess this is the intended/expected behaviour?

The challenge.yml file should only be created once a challenge is done. So there should probably be a ctf challenge finalize kind of command. It can parse the relevant sections that it needs and then as the author for other data.

Feedback suggested that manually generating the yml file was unnecessary work.

Not sure what I was thinking but Alpine tends to introduce weird bugs. I think it's better to not use Alpine and just use things like the Python base images.

Perhaps we can have build system integration with something like https://www.pyinvoke.org/. For example in a reversing challenge we can specify a build script that ctfcli can execute.

I think we may want to consider adding the ability to manage pages from ctfcli. It would help because our editor interface isn't great and this might be a good way to make that easier.

ctf challenge update isn't working properly in some situations for unknown reasons. We need to come up with a better strategy on gathering and updating challenges gathered from multiple locations.

I think writeups should have their own folder instead of being just WRITEUP.md in the root folder. This would make it more organized and also let writeup content not pollute other content.

By doing this it creates a circular reference and overall bad practice.

I am making an "all in one" setup script using docker-compose and kubernetes with some bash and python tying it all together to make a CTFd based pentesting/malware analysis/threat actor analysis/learning project thing-a-ma-bob for my community to learn with but I needed to integrate ctfcli into the overall schema and ended up rewriting it...

I am integrating as many open source challenges as I can into the repository, it is a repository manager with ctfd/git operations that interfaces with the CTFd server. It has a masterlist yaml file that is a representation of the entire challenge set.

I have yet to fully morph the API calls to ctfd into a more abstract schema but I think you should just see it.

I have gotten the masterlist for the challenge repository working, it has read/write methods that turn code to yaml, I am now working on the getter/setter/listing methods, I am going to also get the templating and "build a challenge on the command line" thing going,

The root of the repository is here
https://github.com/mister-hai/sandboxy/tree/master/data/CTFd

the tool itself is here
https://github.com/mister-hai/sandboxy/tree/master/data/CTFd/ctfcli

And the modified open source challenges are in the "challenges" folder

I didnt know how to inform anyone of this properly, I just an trying to contribute to both of our communities and provide a platform for accelerated learning. I am trying very hard to document everything as well as I would have appreciated when I was learning the beginner stuff... I still am learning newb stuff lmao!

Thank you for your time!

We are a monthly event and we already host our challenges in a git repo even when not using CTFd: https://github.com/montrehack/challenges

It would be nice if ctfcli would have detected that it was already in a git tree and wouldn't do anything about it. For now, I'm going to create a distinct repo for our event but I wanted to let you know of that use case.

This tool is a very nice addition to the CTFd ecosystem by the way! Thanks!

With the new documentation site we could have multiple tools documented under one site. Probably worth it to merge some of the documentation together.

We should in theory be able to use CTFd configuration variables in things like challenge descriptions and such. I think this is would be a really useful idea.

Perhaps something like {{ ctf_name }} or {{ ctf_description }}inside of challenge descriptions.

I think we can also have a variables section within the .ctf/config section as well.

It could be useful to have multiple challenges specified in a challenge repo instead of just a single challenge.yml. Or perhaps having multiple challenges inside of a single challenge.yml file. Not sure what the right way to fix this would be.

This needs better documentation on what exactly makes up a challenge. Jumping right into creating an event is a bad flow.

I am still experiencing this error when running the version when I installed this using pip.

This is the pip and pip3 info.

$ pip3 show ctfcli
Name: ctfcli
Version: 0.0.4
Summary: Tool for creating and running Capture The Flag competitions
Home-page: UNKNOWN
Author: Kevin Chung
Author-email: [email protected]
License: Apache 2.0
Location: /home/user/.local/lib/python3.6/site-packages
Requires: requests, colorama, appdirs, fire, cookiecutter, Pygments, click, pyyaml
Required-by: 

And the error I get is as follows.

$ ctf challenge sync Over_The_Top
Found Over_The_Top/challenge.yml
Loaded Over the Top
Syncing Over the Top
Traceback (most recent call last):
  File "/home/user/.local/bin/ctf", line 8, in <module>
    sys.exit(main())
  File "/home/user/.local/lib/python3.6/site-packages/ctfcli/__main__.py", line 74, in main
    fire.Fire(CTFCLI)
  File "/home/user/.local/lib/python3.6/site-packages/fire/core.py", line 138, in Fire
    component_trace = _Fire(component, args, parsed_flag_args, context, name)
  File "/home/user/.local/lib/python3.6/site-packages/fire/core.py", line 471, in _Fire
    target=component.__name__)
  File "/home/user/.local/lib/python3.6/site-packages/fire/core.py", line 675, in _CallAndUpdateTrace
    component = fn(*varargs, **kwargs)
  File "/home/user/.local/lib/python3.6/site-packages/ctfcli/cli/challenges.py", line 134, in sync
    sync_challenge(challenge=challenge)
  File "/home/user/.local/lib/python3.6/site-packages/ctfcli/utils/challenge.py", line 95, in sync_challenge
    files.append(("file", open(f, "rb")))
FileNotFoundError: [Errno 2] No such file or directory: 'src/main.c'

If I make a directory and file in the the folder I am running the file from, the command works just fine. But if I do not have it in that directory, it fails. It should be running it from path that is relative to the challenge.yml location.

Currently I am bypassing this (and recursivly searching folders) using this command.

find . -type d -exec echo '(cd {} && ctf challenge sync)' ';' | sh

We currently generate the README with nothing in it. We can actually generate it from the challenge.yml. We will need some way to promote the filling in of challenge.yml before pushing though as the current flow it's a little easy to miss a step. Maybe with some kind of git hooks.

If you rename a challenge directory or move it it will break the git subtree integration. We may need to look into a mv or rename functionality.

Something to pull down all the data from a given CTFd instance.

Add an option like --no-config or --skip-config to bypass the initial config variable setup.

Getting the following error when trying to use ctfcli. This only began to happen once I upgraded to ctfd 3.4.1 and does not occur on ctfd 3.4.0.

Traceback (most recent call last):
  File "../../utils/ctf", line 11, in <module>
    load_entry_point('ctfcli==0.0.9', 'console_scripts', 'ctf')()
  File "/home/user/.local/lib/python3.8/site-packages/ctfcli/__main__.py", line 78, in main
    fire.Fire(CTFCLI)
  File "/home/user/.local/lib/python3.8/site-packages/fire/core.py", line 138, in Fire
    component_trace = _Fire(component, args, parsed_flag_args, context, name)
  File "/home/user/.local/lib/python3.8/site-packages/fire/core.py", line 466, in _Fire
    component, remaining_args = _CallAndUpdateTrace(
  File "/home/user/.local/lib/python3.8/site-packages/fire/core.py", line 675, in _CallAndUpdateTrace
    component = fn(*varargs, **kwargs)
  File "/home/user/.local/lib/python3.8/site-packages/ctfcli/cli/challenges.py", line 143, in install
    installed_challenges = load_installed_challenges()
  File "/home/user/.local/lib/python3.8/site-packages/ctfcli/utils/challenge.py", line 28, in load_installed_challenges
    return s.get("/api/v1/challenges?view=admin", json=True).json()["data"]
KeyError: 'data'

We should be able to add a Github Action to auto publish to pypi on every release

Not sure if this is intended behavior, but in the previous build of ctfcli, when in a challenge directory, running ctf challenge sync or ctf challenge install would sync/install that challenge

In 0.0.8, it fails silently

Hi!

Maybe i'm missing something but I couldn't find how to use the standard markdown syntax in the description field of the challenge file. If I try to use multiline syntaxe for yaml it seems that every special char is escaped.

Could you provide an example ?
Thanks

If people want to add files to their challenge via ctfcli we should let them tab complete files.

Add a field where we can specify what topics a challenge is about. This should probably also be aligned with CTFd as well. Perhaps we have a set of internal topics.

We need a way to share CTFd configuration between events. It's too easy to forget that a configuration setting needs to be set.

We need a specific templates folder so that we can create new custom templates without having to add them to the main repo. Should be very similar to the plugins subcommand.

Something like:

ctf templates dir
ctf templates install <url>
ctf templates list
ctf templates uninstall <name>

We should have some way to be able to store and use saved copies of images so that we aren't always rebuilding images on deploy.

Essentially if we build once, we should be able to store that image and then reuse it.

add seems kind of like an overloaded term here. I think copying clone makes more sense maybe?

Syncing or installing is an all or nothing procedure. Instead we shouold have switches/options to be able to choose what data to not include during installation.

• Lint that Dockerfiles EXPOSE a port
• Lint that if your challenge spec doesn't provide an image that you don't provide a Dockerfile
• Lint that all files being searched for are provided

When updating challenges with the cli tool, the challenges are set to state "visible", regardless of their previous state.

Expected behavior:
Challenge visibility state stays hidden when being hidden beforehand.

Actual behavior:
Challenge is set to "visible", even if being hidden before.

Add a command to allow you to create the git folder in a different place than the default /home/user/.git

The reasoning is that some of the .git folders have other files in it that may conflict with the install such as secret files. Additionally, I may want to run multiple instances on the same computer.

A decent solution may be creating individual folders in the .git folder for each individual project. IE /home/user/.git/ctf1 /home/user/.git/ctf2 etc.

There should be a blank challenge template to make it easier to try out ctfcli

When we get errors from the API or otherwise get a bad response we need to print out the response to make it easier to debug.

The spec does not have the following fields for challenges having type dynamic:

• Initial Value
• Decay Limit
• Final Value

Hence, you can only set challenges of standard type.