creative-writing-modes
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is composed entirely of instructional markdown files designed to guide an AI through creative writing workflows. No executable code, scripts, or privilege-escalating commands are present in the provided files.
- [INDIRECT_PROMPT_INJECTION]: The skill instructions involve reading external project data (scene briefs, outlines, style files, character states, and critique synthesis) to inform the generated prose. While this represents a surface for indirect prompt injection, the risk is negligible as the skill lacks high-privilege capabilities such as file system writing, network exfiltration, or shell command execution.
- Ingestion points: Instructions in
resources/prose-modes.mdreference reading scene briefs, outlines, style files, canon anchors, and critique synthesis documents. - Boundary markers: The instructions do not define explicit delimiters or 'ignore instructions' wrappers for external data.
- Capability inventory: The skill is limited to text generation; no tools, network operations, or subprocess executions are invoked.
- Sanitization: The skill does not implement specific sanitization or filtering for the data it processes.
Audit Metadata