google-calendar
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
@membranehq/clipackage from the public NPM registry. This is a vendor-provided tool used for authentication and API communication. - [COMMAND_EXECUTION]: The instructions involve executing several shell commands using the
membraneCLI to log in, establish connections, and perform calendar operations. - [INDIRECT_PROMPT_INJECTION]: The skill acts as a bridge to external data from Google Calendar, creating an attack surface for indirect prompt injection.
- Ingestion points: Data retrieved from
membrane action list,membrane action run, andmembrane request(e.g., event descriptions, calendar names) is ingested into the agent's context (SKILL.md). - Boundary markers: The skill lacks explicit instructions or delimiters to warn the agent about potentially malicious content within the external data.
- Capability inventory: The agent has the capability to execute shell commands and perform network operations via the provided CLI (SKILL.md).
- Sanitization: No sanitization or content filtering is implemented for the data fetched from the external API.
Audit Metadata