easy-notion-mcp-integration

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the "easy-notion-mcp" package from the public npm registry using "npx" and provides a repository URL for source code cloning ("https://github.com/Grey-Iris/easy-notion-mcp.git").
  • [COMMAND_EXECUTION]: Instructs users to execute standard build and deployment commands such as "npm install", "npm run build", and "npm start" to run the MCP server.
  • [DATA_EXFILTRATION]: The skill facilitates data movement between the local environment and the Notion API, which is its primary function. It contains security measures like "NOTION_MCP_WORKSPACE_ROOT" to limit the "upload_file" tool's access to a predefined workspace directory, preventing directory traversal.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted data from external sources.
  • Ingestion points: Tools like "read_page", "query_database", and "search_notion" fetch markdown content from Notion.
  • Boundary markers: Implements a "NOTION_TRUST_CONTENT" toggle (enabled by default) that prepends a warning prefix to content read from Notion, alerting the agent to potentially untrusted instructions.
  • Capability inventory: Includes capabilities to modify Notion data ("update_page", "delete_page"), create new entries ("create_page", "add_database_entry"), and read local files ("upload_file").
  • Sanitization: The documentation explicitly mentions content sanitization and URL validation as built-in features.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 04:48 AM
Security Audit — agent-trust-hub — easy-notion-mcp-integration