aepd-es
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs tools and connects to an MCP server hosted on the vendor's domain (
unbrowse.ai). As these resources belong to the skill's author, they are considered standard operational dependencies. - [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes data from external web pages (
aepd.es), creating a surface for indirect prompt injection where malicious instructions embedded in the website could attempt to influence the agent's behavior. - Ingestion points: Content is retrieved from the AEPD website via
aepd_es__get_buscador,aepd_es__get_preguntas_frecuentes_buscador, andaepd_es__read_pagetools. - Boundary markers: No specific delimiters or instructions are provided to the agent to treat the fetched content as untrusted data.
- Capability inventory: The tools are read-only and return page titles, text, and links to the agent context.
- Sanitization: There is no evidence of content filtering or sanitization performed on the raw text before it is presented to the agent.
Audit Metadata