malaysiakini
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches MCP configurations and tool definitions from the vendor's official domain at unbrowse.ai.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from malaysiakini.com which could potentially contain malicious instructions.
- [INGESTION_POINTS]: The 'query' parameter in 'malaysiakini_com__get_en_search' and the 'path' parameter in 'malaysiakini_com__read_page' (SKILL.md).
- [BOUNDARY_MARKERS]: The skill does not define specific delimiters or instructions to ignore embedded instructions in the fetched data.
- [CAPABILITY_INVENTORY]: The skill uses tool calls that perform network requests to the vendor's infrastructure to fetch content.
- [SANITIZATION]: No explicit sanitization or filtering of the external content is described.
Audit Metadata