web-page-workflow
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The file
prototype-restoration.mdinstructs the agent to run a local script using the commandnode scripts/check-app-ready.mjs /prototypes/[page-name]. This execution depends on arguments derived from the page restoration process. - [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it processes untrusted data from external URLs and uses it to generate executable React code and documentation.
- Ingestion points: External web content is ingested through
get_page_markdown,get_page_theme, and Firecrawl MCP tools. - Boundary markers: None identified. The instructions do not provide delimiters or specific warnings to the agent to ignore instructions that might be embedded in the scraped content.
- Capability inventory: The skill has the capability to write files to the
src/andtemp/directories and execute shell commands vianode. - Sanitization: There are no instructions provided to sanitize, escape, or validate the content retrieved from web pages before it is used for code generation.
Audit Metadata