redactor
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted book content and is instructed to edit source files in place. This creates a surface for indirect prompt injection where malicious instructions embedded in the source text could influence the agent's behavior.
- Ingestion points: Input content provided by the user, such as Markdown files or quoted excerpts, as processed in
SKILL.md. - Boundary markers: There are no specific instructions for the agent to use delimiters or ignore embedded instructions when reading the source content.
- Capability inventory: The skill has the ability to perform file-write operations, specifically 'edit that source in place'.
- Sanitization: No sanitization or validation of the input content is mentioned in the skill instructions.
Audit Metadata