flash-fiction
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill is strictly documentation-based, consisting of Markdown diagnostic states, checklists, and writing examples. It does not include scripts, binaries, or configuration files that execute commands.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-provided fiction for evaluation. While this creates an ingestion surface for potential prompt injection, the risk is negligible due to the lack of side-effect capabilities.
- Ingestion points: Story content provided by the user for analysis.
- Boundary markers: None defined in the diagnostic instructions.
- Capability inventory: No file system operations, network access, or shell command execution capabilities are utilized or requested by the skill.
- Sanitization: None present; the skill relies on standard model reasoning for evaluation.
Audit Metadata