novel-evaluator

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection through the processing of untrusted story text.
  • Ingestion points: The skill accepts 'Complete story text or story outline' as input for evaluation (SKILL.md).
  • Boundary markers: The instructions do not use delimiters or provide warnings to the agent to ignore instructions embedded within the user-provided story content.
  • Capability inventory: The skill is restricted to the 'Read' tool as specified in the frontmatter; no network, file-write, or shell execution capabilities are present.
  • Sanitization: There is no evidence of input validation or content filtering described in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 10:24 PM
Security Audit — agent-trust-hub — novel-evaluator