The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its core function of fetching and analyzing untrusted web content. Ingestion points: SKILL.md instructs the agent to fetch and read rendered pages from provided URLs. Boundary markers: The instructions lack specific delimiters for the external content. Capability inventory: The agent is limited to text analysis and report generation based on the data. Sanitization: No content filtering or sanitization is mentioned in the audit process.
[NO_CODE]: The skill contains no executable code, scripts, or configuration files that initiate external processes. All provided files are Markdown references or JSON-LD templates.

eeat-audit