The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches the xfetch-cli Node.js package from the LXGIC-Studios/xfetch GitHub repository.
[COMMAND_EXECUTION]: The skill relies on executing the xfetch binary to perform data retrieval and search operations on X/Twitter.
[CREDENTIALS_UNSAFE]: The skill requires users to manage and set sensitive session cookies (auth_token and ct0) through CLI commands. Providing these as command-line arguments can result in their exposure in shell history or system process logs.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core function of fetching untrusted content from X/Twitter. \n
Ingestion points: Untrusted content is ingested from the external platform via xfetch tweets, xfetch search, and xfetch thread commands. \n
Boundary markers: No delimiters or boundary markers are present in the skill instructions to isolate fetched data from agent instructions. \n
Capability inventory: The skill possesses the capability to execute system commands through the xfetch utility. \n
Sanitization: There is no evidence of sanitization, filtering, or validation of the retrieved tweet content before it is processed.

xfetch