character-naming
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes local TypeScript scripts (
character-name.tsandcast-tracker.ts) via the Deno runtime to perform its core functions. These scripts are executed with specific flags (--allow-read,--allow-write) to manage local character data files.\n- [PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection because user-provided strings (names, roles, project titles) are passed as command-line arguments to Deno scripts.\n - Ingestion points: Character names, roles, and setting details provided by the user.\n
- Boundary markers: No explicit delimiters or sanitization instructions are defined for shell command arguments.\n
- Capability inventory: Subprocess execution via Deno and local filesystem write access.\n
- Sanitization: Not defined in the provided instructions.
Audit Metadata