riksdag-regering-mcp

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process large volumes of external data from Swedish government sources (data.riksdagen.se, regeringen.se, g0v.se), which creates a surface for indirect prompt injection.
  • Ingestion points: 32 specialized tools fetch documents, motions, reports, and speeches from external Swedish government APIs.
  • Boundary markers: None identified in the skill instructions to separate external data from system instructions.
  • Capability inventory: The skill provides extensive data retrieval and search capabilities. While it does not explicitly contain shell execution or file-write commands, the retrieved content is intended for analysis and summarization by the agent.
  • Sanitization: No explicit sanitization or validation of the fetched external content is described.
  • [EXTERNAL_DOWNLOADS]: The documentation references an external package and a production endpoint for the MCP server.
  • Evidence: Recommends installing the riksdag-regering-mcp package from NPM and connects to the production endpoint at https://riksdag-regering-ai.onrender.com/mcp.
  • Context: These resources are provided by the skill author (hack23) and are consistent with the skill's stated purpose of providing political data access.
  • [COMMAND_EXECUTION]: The skill provides instructions for local installation and execution of the MCP server.
  • Evidence: npm install -g riksdag-regering-mcp and riksdag-regering-mcp are provided as setup commands.
  • Context: These are standard deployment instructions for the associated tool and are presented for development purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 11:51 PM