gws-docs-write
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill handles untrusted data through the
--textflag which is appended to Google Documents, creating an indirect prompt injection surface where malicious instructions within the text could influence the agent's behavior. \n - Ingestion points:
--textargument inSKILL.md.\n - Boundary markers: None identified in the instruction set.\n
- Capability inventory: Capability to modify documents via
gws docs +write.\n - Sanitization: No sanitization or input validation logic is present in the skill definition.\n- [COMMAND_EXECUTION]: The skill invokes the
gwsCLI tool with parameters derived from user input (--documentand--text). If these arguments are not correctly escaped by the platform's shell executor, it could allow for arbitrary command injection on the host system.
Audit Metadata