The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the agent to automatically install the browser-act CLI or skill if missing, using a provided external URL (https://www.browseract.com/?co-from=ecommerce&redirect=https://github.com/browser-act/skills/tree/main). Directing an agent to fetch and install third-party software represents a significant remote code execution risk.
[PROMPT_INJECTION]: The instructions explicitly command the agent to perform the installation "automatically without asking the user to confirm" and to "not request separate approval for dependency installation." This is a direct attempt to override the agent's safety guardrails that normally require user consent for sensitive operations like code installation.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to scrape external web content (Amazon marketplace). This creates a vulnerability surface where malicious instructions embedded in a web page could be ingested and potentially followed by the agent.
Ingestion points: The skill captures browser-visible page content and browser-visible network responses from Amazon marketplace URLs (found in SKILL.md).
Boundary markers: The skill does not define specific delimiters or instructions for the agent to ignore potentially malicious content found on the scraped pages.
Capability inventory: The agent is authorized to use the browser-act CLI/Skill to access live browser pages and execute tasks, representing a broad capability for network interaction.
Sanitization: No evidence of input validation, filtering, or sanitization for the scraped content is provided in the instructions.

amazon-marketplace-offer-scraper