grist
Pass
Audited by Gen Agent Trust Hub on Apr 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
membraneCLI tool to perform various operations including searching for connectors, managing connections, and executing API actions against Grist. - [EXTERNAL_DOWNLOADS]: Instructs the user to install the
@membranehq/clipackage globally via npm. This package is the official tool provided by the skill's author to facilitate secure communication with external services. - [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection because it retrieves and processes data from external Grist tables and documents.
- Ingestion points: Data enters the agent's context through actions like
list-records,run-sql-query, andget-documentdescribed inSKILL.md. - Boundary markers: There are no explicit instructions or delimiters defined to separate untrusted data from agent instructions.
- Capability inventory: The skill possesses significant capabilities through the
membraneCLI, including the ability to create, update, and delete records, tables, and workspaces. - Sanitization: No sanitization or validation logic is specified for the data retrieved from Grist before it is interpreted by the agent.
Audit Metadata