character-development
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a framework for reading from and writing to character data files, which constitutes a data ingestion surface.\n
- Ingestion points: The skill reads character definitions from
characters/CHARACTER.md,style/CHARACTER_VOICE.md, andcharacters/RELATIONSHIP.md(SKILL.md).\n - Boundary markers: No specific boundary markers or instruction-protection delimiters are specified for the ingested content.\n
- Capability inventory: The skill contains no subprocess calls, network operations, or privilege escalation capabilities. Its functions are restricted to standard agent text processing.\n
- Sanitization: No sanitization or input validation is performed on the data read from character files.\n- [SAFE]: No malicious patterns such as prompt injection, data exfiltration, or obfuscation were detected in the instructions or metadata.\n- [SAFE]: The skill does not perform any remote code execution or external package installations.
Audit Metadata