The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides an installation command that pipes a remote shell script to bash (curl -fsSL https://raw.githubusercontent.com/xdevplatform/xurl/main/install.sh | bash). This script is sourced from the official xdevplatform GitHub repository.
[EXTERNAL_DOWNLOADS]: Installation instructions reference downloads from GitHub, Homebrew, npm, and Go package registries.
[COMMAND_EXECUTION]: The skill invokes the xurl CLI tool and a python3 script via heredoc to process social media data.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its X Article reading fallback.
Ingestion points: The Python script in SKILL.md fetches content from external mirror APIs api.fxtwitter.com and api.vxtwitter.com.
Boundary markers: Absent; the script prints data directly to the agent's context.
Capability inventory: The skill includes the ability to post, delete, and send DMs on X, as well as general shell command execution.
Sanitization: Absent; the retrieved text is not filtered for potential injection patterns.
[CREDENTIALS_SAFE]: The skill includes extensive warnings and rules to prevent the agent from accessing or leaking the ~/.xurl configuration file or handling sensitive API keys directly.

xurl