creator-shortlist
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The README.md file directs users to install additional capabilities using 'npx skills add unifapi-agent/agents'. This command downloads and executes code from the vendor's own repository to set up the environment.
- [DATA_EXFILTRATION]: The skill is instructed to access and read local files, specifically '.agents/product-marketing.md' and '.claude/product-marketing.md', to retrieve campaign briefs. While this is the intended functionality for marketing research, it involves reading from the local file system.
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its core function of processing untrusted data from external sources.
- Ingestion points: The skill reads user-generated content from social media platforms (X, TikTok, YouTube, Instagram), including profile metadata, tweets, and video descriptions, via the UnifAPI MCP server.
- Boundary markers: There are no explicit instructions or delimiters used to separate untrusted external data from the agent's core instructions.
- Capability inventory: The skill has the ability to read local files and access the network through the vendor's API. It does not have access to arbitrary command execution, system-level modifications, or file-writing capabilities.
- Sanitization: The skill does not implement validation or sanitization for the social media data it ingests before processing it.
Audit Metadata