create-viral-content

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Potential for Indirect Prompt Injection (Category 8).
  • Ingestion points: The skill ingests user drafts and social media content through instructions in SKILL.md and various platform templates in resources/platform-templates.md.
  • Boundary markers: No explicit delimiters or boundary markers (such as XML tags or unique string wrappers) are used to separate untrusted input content from the agent's instructions.
  • Capability inventory: The skill allows the agent to generate new text based on input, perform multi-step refinement loops, and invoke secondary skills such as humanize-writing (noted in resources/humanize-integration.md).
  • Sanitization: No input data sanitization or validation logic is present to filter out potential malicious instructions embedded within the content being refined.
  • [PROMPT_INJECTION]: Metadata inconsistency (Category 7). The author is listed as ice-ninja in the skill files, but the system context identifies the author as aAAaqwq. This discrepancy in metadata can complicate identity verification and trust assessment.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 06:59 AM