The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses several Python scripts to perform specific verification tasks, such as DOI lookups and URL reachability checks. These scripts are self-contained and use standard library functions.
[EXTERNAL_DOWNLOADS]: To verify literature, the skill connects to well-known academic databases and search engines, including CrossRef, CNKI, and Baidu Scholar. These connections are necessary for fetching the metadata used in the verification process.
[DYNAMIC_EXECUTION]: The skill provides the agent with JavaScript templates to extract data from academic search pages using browser automation tools. This dynamic execution is scoped to data extraction from specific DOM elements.
[DATA_EXFILTRATION]: No evidence of unauthorized data transfer was found. The skill only retrieves academic metadata and displays it to the agent/user, with no access to sensitive local files or environment secrets.
[PROMPT_INJECTION]: The skill processes external text from web sources (paper titles, meta tags), which presents a potential surface for indirect prompt injection. This risk is minimized by the skill's focused logic and basic HTML sanitization in the extraction scripts.

literature-verifier