write-without-slop
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructions (in
SKILL.mdunder 'Whose voice?') direct the agent to ingest and analyze user-supplied text samples to fix on specific signals like vocabulary and sentence length. This creates a surface for indirect prompt injection where malicious instructions could be embedded in the analyzed samples to influence the agent's behavior. - Ingestion points:
SKILL.mddescribes taking 'voice samples' from nearby chapters, earlier posts, or surrounding documentation to emulate a specific tone. - Boundary markers: There are no instructions for the agent to treat the samples as untrusted or to use delimiters to separate stylistic data from instructional content.
- Capability inventory: The skill is primarily focused on text generation and does not configure or invoke high-privilege tools such as file-system writes, network operations, or shell command execution.
- Sanitization: No sanitization or filtering logic is specified for the input samples.
Audit Metadata