mmk-notion-database
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
mmkcommand-line tool, a vendor-specific resource frommagic-meal-kits, to perform Notion database operations. These operations include schema retrieval, querying with filters, and bulk data manipulation. - [DATA_EXFILTRATION]: The skill supports reading data from the local filesystem via the
--fileflag in theinsert,update,upsert, anddeletecommands. This allows the agent to read local JSON files and transmit their contents to Notion's cloud infrastructure. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it retrieves untrusted content from external Notion databases.
- Ingestion points: Data is ingested through the
queryandschemacommands inSKILL.md. - Boundary markers: The instructions do not define delimiters or specific safety instructions to isolate data retrieved from Notion.
- Capability inventory: The skill has capabilities to modify Notion data (
update,delete,upsert) and trigger AI workflows (ai-summary) based on the ingested content. - Sanitization: There is no evidence of content sanitization or validation performed on the records retrieved from external databases before they are processed by the agent.
Audit Metadata