notemdpro-content-generator
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches web content via search providers (Tavily, DuckDuckGo) and direct URL fetching to provide context for content generation.
- [COMMAND_EXECUTION]: Performs automated file system operations including batch markdown generation and file-based syntax repair for Mermaid diagrams.
- [PROMPT_INJECTION]: Risk of indirect prompt injection through external data ingestion.
- Ingestion points: Web content retrieved via
_performResearchandfetchContentFromUrlinSKILL.md. - Boundary markers: The prompt template in
SKILL.mduses a{RESEARCH_CONTEXT_SECTION}placeholder but lacks explicit instructions to ignore potentially malicious directions within that context. - Capability inventory: The skill has file system write access (
batchGenerateContentForTitles,fixMermaidSyntaxInFile) and network read capabilities. - Sanitization: No security-focused sanitization is documented; only formatting-related cleanup (
cleanupLatexDelimiters,refineMermaidBlocks) is performed.
Audit Metadata