coo-social-media
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill integrates with Playwright MCP tools to automate social media operations. The navigation targets are well-known, official creator platforms (creator.xiaohongshu.com, creator.douyin.com, mp.weixin.qq.com), which is consistent with the skill's stated purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it retrieves and processes untrusted data from external social media websites.
- Ingestion points: Data is ingested from external web pages via
mcp__playwright__playwright_get_visible_textandmcp__playwright__playwright_get_visible_htmlinSKILL.md. - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the retrieved social media content.
- Capability inventory: The agent has access to Playwright tools enabling page navigation, element interaction (click/fill), and script evaluation.
- Sanitization: There is no explicit sanitization or filtering logic applied to the content extracted from social media pages before it is returned to the agent context.
Audit Metadata