The Agent Skills Directory

[NO_CODE] (SAFE): The analyzed skill contains only markdown instructions and metadata. No scripts (.py, .js, .sh), binaries, or configuration files for package managers were found.
[PROMPT_INJECTION] (SAFE): The instructions define a professional persona and a structured workflow for fact-checking. There are no attempts to bypass safety filters, extract system prompts, or disregard previous instructions.
[DATA_EXFILTRATION] (SAFE): There are no network-capable commands (curl, wget, fetch) or references to sensitive file paths (~/.ssh, ~/.aws) that would facilitate data theft.
[EXTERNAL_DOWNLOADS] (SAFE): No external dependencies or remote script executions are requested or performed.
[INDIRECT_PROMPT_INJECTION] (LOW): The skill is intended to process untrusted user data (claims to be verified). While this is an attack surface, the skill lacks any tools or executable capabilities (like file writing or system commands) that an attacker could exploit through a malicious claim. The risk is limited to potential bias or misleading output in the text response.

fact-checker