Research Topic
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE] (SAFE): The skill is composed strictly of markdown templates and process instructions for an AI agent. No shell commands, scripts, or system-level interactions are present.
- [Indirect Prompt Injection] (LOW): 1. Ingestion points: External codebase files located in pkg/, docs/, and examples/ directories. 2. Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are used for ingested content. 3. Capability inventory: The skill has no capabilities for code execution, file modification, or network access. 4. Sanitization: No sanitization logic is defined for the content extracted from the codebase. Severity is LOW because it is restricted to reasoning and display only.
Audit Metadata