The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes headless Chrome browser processes via the Selenium library in hidden_text_detection.py and keyword_stuffing_detection.py to analyze computed styles and rendered page content.
[EXTERNAL_DOWNLOADS]: The skill performs network requests to fetch data from external URLs provided for auditing in cloaking_detection.py and sneaky_redirect_detection.py. It also utilizes webdriver-manager to download and manage browser binaries.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and process untrusted data from the web.
Ingestion points: External HTML and text content are retrieved from audited URLs in cloaking_detection.py, hidden_text_detection.py, keyword_stuffing_detection.py, and sneaky_redirect_detection.py.
Boundary markers: There are no explicit delimiters or system instructions used to encapsulate the ingested content before it is processed by the agent.
Capability inventory: The skill possesses capabilities including network access (requests) and the ability to spawn browser subprocesses (Selenium).
Sanitization: The skill employs BeautifulSoup's get_text() and standard regular expression cleaning to strip HTML tags and normalize whitespace. This sanitization reduces the risk of structural injection but does not fully prevent semantic instructions embedded in the audited content from influencing the agent's logic.

seo-engine