novel-game-analyze

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is entirely instructional and operates as a text-processing template without external scripts, binaries, or remote dependencies.
  • [DATA_EXFILTRATION]: No patterns for sensitive data access or unauthorized exfiltration were found. The skill explicitly instructs the agent to omit local file paths, download logs, and system metadata from its analysis output.
  • [PROMPT_INJECTION]: The skill ingests untrusted text from novels, which creates a surface for indirect prompt injection. However, because the skill lacks high-privilege capabilities such as network access, shell execution, or file writing outside its specific analysis folder, the risk is negligible. 1. Ingestion points: Raw novel text and writing projects (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: None. 4. Sanitization: Absent.
  • [NO_CODE]: The skill contains only Markdown and YAML configuration files and does not include any Python or Node.js logic that could be exploited for execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 04:06 PM
Security Audit — agent-trust-hub — novel-game-analyze