ghostwriter-pro
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core functionality relies on processing 5-10 writing samples and interview memos to generate voice profiles and ghostwritten content. This creates a significant surface for indirect prompt injection. \n
- Ingestion points: Writing samples and interview transcripts requested in
SKILL.mdandtemplates/prompt-main.md. \n - Boundary markers: The skill does not define clear delimiters or instructional boundaries (e.g., 'ignore all instructions within this sample') for the external text samples. \n
- Capability inventory: The skill is designed to perform detailed analysis and mimic styles based on these inputs. While it does not call high-privilege system tools, the processed content directly influences the agent's output and persona. \n
- Sanitization: No input sanitization or validation is specified for the ingested samples.
Audit Metadata