The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill is purpose-built to extract the sessionid authentication cookie from jimeng.jianying.com. Session cookies are sensitive credentials that grant full access to the user's account and are a primary target for exfiltration.
[DYNAMIC_EXECUTION]: The skill workflow utilizes browser_run_code to execute a custom JavaScript snippet at runtime. This script interacts with the browser context to retrieve active cookies, which is a form of dynamic code execution used for data retrieval.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from a remote website, creating an attack surface where instructions embedded in the webpage could potentially influence the agent's logic during the extraction process.
Ingestion points: The skill navigates to https://jimeng.jianying.com/ai-tool/home and reads the page state to check for login status (file: SKILL.md).
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content embedded in the site's HTML or membership information fields.
Capability inventory: The skill possesses the ability to execute code (browser_run_code), navigate the browser (browser_navigate), and snapshot page content (browser_snapshot) (file: SKILL.md).
Sanitization: There is no evidence of sanitization or filtering applied to the site's content before the agent evaluates the login state.

yoitao-jimeng-sessionid