The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the dig and curl command-line utilities via Python's subprocess.run. These tools are used for their intended purposes of DNS resolution and fetching website source code. The implementation uses argument lists rather than shell strings, which effectively prevents shell injection attacks from untrusted domain inputs.
[EXTERNAL_DOWNLOADS]: The script performs network operations to retrieve public DNS records and website HTML content from user-specified domains. It also communicates with the vendor's API at app.gooseworks.ai to perform deeper technology profiling. All external communications are consistent with the skill's stated purpose.
[PROMPT_INJECTION]: The skill ingests untrusted data from third-party websites. This creates a surface for indirect prompt injection if a target website contains malicious instructions. However, the script mitigates this risk by using specific regular expression patterns to extract known technology signatures (such as tracking IDs) rather than including the raw HTML content in the agent's context.

tech-stack-teardown