mcp-server-integration
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads and installs official Model Context Protocol server packages from the npm registry, including
@modelcontextprotocol/server-filesystem,@modelcontextprotocol/server-memory,@modelcontextprotocol/server-sequential-thinking, and@playwright/mcp. - [COMMAND_EXECUTION]: Defines local commands to initiate MCP servers, such as
mcp-server-filesystemandmcp-server-memory, as well as package execution vianpxfor the Playwright integration. - [PROMPT_INJECTION]: Contains a vulnerability surface for Indirect Prompt Injection (Category 8).
- Ingestion points: Processes external data from web pages via
playwright-browser_navigate, local files viafilesystem-read_text_file, and GitHub repository content. - Boundary markers: No explicit delimiters or instructions to disregard embedded instructions are defined for the data processed by these tools.
- Capability inventory: Includes high-impact capabilities such as writing to the filesystem (
filesystem-write_file), modifying GitHub repositories (github-create_issue), and browser automation. - Sanitization: The configuration does not specify any sanitization or validation routines for data fetched from untrusted external sources.
Audit Metadata