The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to install dependencies (pip install -r requirements.txt) and run included Python scripts (scripts/fetch_page.py, scripts/parse_html.py) for data processing.
[EXTERNAL_DOWNLOADS]: The skill fetches HTML content from user-provided URLs using the requests library. While this is the intended purpose, it involves downloading untrusted content from the public internet.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8). It processes untrusted data from external sources that could contain instructions to hijack the agent's behavior.
Ingestion points: External HTML content fetched by scripts/fetch_page.py and parsed by scripts/parse_html.py (specifically titles, meta tags, and alt text).
Boundary markers: Absent. The analysis phase (Phase 2) does not define clear delimiters or instructions for the agent to ignore commands within the analyzed content.
Capability inventory: The agent has the ability to execute shell commands and make network requests.
Sanitization: No sanitization or filtering is performed on the extracted HTML content before it is processed by the AI context.

seo-page