gws-people
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a required system binary
gwsto perform all API operations. This is the intended design for interacting with Google Workspace services via the command line. - [INDIRECT_PROMPT_INJECTION]: The skill provides the ability to read and search contact data which may contain untrusted information that could influence the agent's behavior.
- Ingestion points: Data is ingested through API methods such as
people.get,people.listDirectoryPeople,people.searchContacts, andotherContacts.listdescribed inSKILL.md. - Boundary markers: There are no explicit delimiters or specific 'ignore instructions' warnings defined in the skill instructions to separate contact data from agent instructions.
- Capability inventory: The skill possesses the capability to create, update, and delete contact groups and individual contacts using the
gwstool, as listed in theAPI Resourcessection ofSKILL.md. - Sanitization: No content sanitization or validation of the retrieved contact fields is specified before the data is processed by the agent.
Audit Metadata