anti-ai-slop
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill employs authoritative and coercive directives intended to override the agent's default behavior and other instructions. Evidence includes the use of terms like 'NON-NEGOTIABLE', 'MUST follow', 'Never skip it', and 'overrides stylistic preferences'.
- [PROMPT_INJECTION]: The skill includes deceptive metadata in the YAML frontmatter, specifically a future date (2026-06-07) and a reference to a non-existent 'EVIDENCE-AUDIT.md' file, likely intended to manufacture a false sense of authority or 'official' status.
- [PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface by requiring the real-time processing of all agent output without boundary markers. 1. Ingestion points: Agent output generation (SKILL.md). 2. Boundary markers: Absent; there are no instructions to ignore malicious directives within the content being filtered. 3. Capability inventory: Real-time filtering and modification of all output types (text, code, UI). 4. Sanitization: Absent.
- [SAFE]: The skill references the author's website 'techguypeter.com', which is a recognized vendor resource.
Audit Metadata