The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to clone a third-party repository from https://github.com/defparam/smuggler. This external source is not a recognized trusted organization.
[COMMAND_EXECUTION]: The skill executes multiple commands that run third-party code, including pip3 install -r requirements.txt to install unverified dependencies and python3 smuggler.py to run the downloaded script.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data in the form of URL lists and piped input to perform network operations.
Ingestion points: The skill reads from user-provided files such as urls.txt and endpoints.txt, or via standard input piping from tools like subfinder and httpx.
Boundary markers: There are no explicit instructions or delimiters to isolate or ignore potentially malicious instructions embedded in the processed target data.
Capability inventory: The skill utilizes network socket operations to send HTTP mutations via smuggler.py and writes results to local files using the -l flag.
Sanitization: No sanitization or validation of the input URLs is mentioned in the instructions before they are passed to the execution script.

smuggler