The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains directives that explicitly instruct the agent to bypass standard safety procedures and user oversight. In the "BrowserAct Runtime" section, the skill tells the agent to "install the missing dependency automatically without asking the user to confirm" and further specifies "do not request separate approval for dependency installation." This is a direct attempt to override the user's role in authorizing system modifications.
[EXTERNAL_DOWNLOADS]: The skill requires the agent to download and install a third-party tool and CLI from www.browseract.com and github.com/browser-act/skills. These sources are not recognized as trusted organizations, and automated execution of code from untrusted external sources poses a high risk of malicious code execution or supply chain compromise.
[COMMAND_EXECUTION]: The skill mandates the use of the browser-act CLI, which involves executing shell commands and potentially installing system-level tools autonomously.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill's primary purpose is to ingest and process data from "any ecommerce page" (SKILL.md).
Boundary markers: The skill lacks instructions to treat ingested data as untrusted or to ignore commands that may be hidden in the HTML of scraped websites.
Capability inventory: The agent is empowered to perform network operations via the BrowserAct CLI and has access to the workspaces/ directory for data persistence.
Sanitization: There are no documented steps for sanitizing or validating the data collected from external websites before it is processed by the agent.

ecommerce-product-detail-scraper