Character Continuity
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were identified. The skill is instructional and limits its operations to the defined media API.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data which serves as a potential ingestion point for indirect instructions.
- Ingestion points: Reference sets, identity anchors, and approved media files (SKILL.md).
- Boundary markers: The "Safety and authorization gate" section specifies verification steps for likenesses (SKILL.md).
- Capability inventory: Capabilities are limited to media generation, editing, and uploading via the muapi connection (SKILL.md).
- Sanitization: The skill requires authorization verification, provenance tracking via manifests, and labeling of synthetic outputs (SKILL.md).
- [COMMAND_EXECUTION]: All operations are restricted to the muapi connection; no shell commands or system-level executions were found.
- [DATA_EXPOSURE]: The skill emphasizes the use of an asset manifest for URLs and IDs to prevent sensitive information from being embedded in plain-text prose.
Audit Metadata