self-media-platform-copywriting
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of instructional markdown files and configuration files for a copywriting agent. Its logic focuses on content transformation, platform-specific formatting, and stylistic consistency.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external inputs described as "evidence packages" and "creative briefs" to generate social media posts. This creates an inherent surface for indirect prompt injection where malicious instructions could be embedded in the source material.
- Ingestion Points:
SKILL.md(processes external evidence, briefs, and themes). - Boundary Markers: The instructions do not define specific delimiters or "ignore" prefixes to isolate the untrusted evidence content from the agent's core instructions.
- Capability Inventory: The skill generates text and visual prompts and references other internal publisher skills (
self-media-wechat-publisher,self-media-short-video) for downstream tasks. - Sanitization: No sanitization or validation of the input evidence text is specified before it is interpolated into the generation workflow.
Audit Metadata