The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the way it processes external data.
Ingestion points: The skill uses the WebFetch tool to retrieve content from user-provided URLs for analysis (SKILL.md).
Boundary markers: There are no explicit delimiters or instructions to the agent to ignore potentially malicious commands embedded within the retrieved HTML content.
Capability inventory: The skill has access to a broad range of powerful tools, including Bash, WebFetch, Read, Grep, and Glob (SKILL.md).
Sanitization: No sanitization or validation logic is defined to filter the content of the web pages before it is processed by the AI agent.
[COMMAND_EXECUTION]: The skill requests access to the Bash shell tool in its metadata (SKILL.md). Although no specific scripts are provided in the skill package, granting shell access in a skill that also ingests untrusted content from the web creates a potential execution vector if the agent is manipulated by instructions found on a scanned page.
[EXTERNAL_DOWNLOADS]: The skill is designed to download content from arbitrary external websites provided by the user via the WebFetch tool (SKILL.md).
[DATA_EXFILTRATION]: The skill includes an optional integration with DataForSEO MCP tools, which involves sending page data and search terms to an external third-party service (SKILL.md).

seo-page