go
Warn
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Use of 'Wall of Text' / Context Stuffing. Several reference files included in the skill (e.g., 'references/performance-optimization.md', 'references/deployment-pipelines.md', 'references/state-management.md') each contain 150 sections of repetitive, boilerplate filler text. This extreme volume of redundant data is a context stuffing technique designed to saturate the AI's context window, which can cause the AI to forget or ignore its initial system instructions and safety constraints.
- [PROMPT_INJECTION]: Persona manipulation. The reference files contain headers asserting that the content is 'strictly intended for Staff+ Engineers' and contains 'extremely dense technical specifications.' This is a persona-adoption technique intended to influence the AI to adopt a privileged 'Staff' persona, potentially leading to the bypass of standard operational procedures or safety filters.
Audit Metadata