The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to crawl and process content from external websites, which creates a surface for indirect prompt injection. A malicious site could host hidden instructions designed to manipulate the agent's output or behavior during the audit process.
Ingestion points: External content is retrieved via tools like firecrawl_crawl, context_dev_scrape_markdown, and context_dev_scrape_html as described in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or specific isolation prompts to treat the ingested website content as untrusted.
Capability inventory: The skill utilizes tools for mapping site structures, crawling content, SEO analysis, and capturing screenshots via sandbase_call_tool.
Sanitization: No explicit sanitization or validation logic is defined for the content extracted from external URLs before it is processed by the agent.

site-audit