battle-choreography
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of instructional Markdown for creative writing and logical consistency auditing. No security vulnerabilities or malicious patterns were detected.
- [NO_CODE]: There are no executable scripts, shell commands, or external dependencies included in the skill files.
- [INDIRECT_PROMPT_INJECTION]: The skill acts as a reasoning framework for user-supplied narrative data. Evidence Chain: 1. Ingestion points: Narrative scenarios processed via instructions in SKILL.md; 2. Boundary markers: Absent; 3. Capability inventory: Local file-write to "battles/BATTLE.md"; 4. Sanitization: Absent. The risk is considered safe as the skill lacks network or privileged command capabilities.
Audit Metadata