local-axure-workflow
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local script
node scripts/check-app-ready.mjsto verify the generated code, which involves local process execution based on internal skill logic. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted content from local files like
content.mdandsitemap.jsonwithout sanitization. - Ingestion points: Data is ingested from local project files including
sitemap.json,content.md, andtheme.json. - Boundary markers: No explicit markers are used to isolate untrusted content from the agent's instructions.
- Capability inventory: The agent has permissions to read/write files, list directories, and execute shell commands via node.
- Sanitization: No evidence of sanitization or instruction-stripping for the processed text content is provided.
- [EXTERNAL_DOWNLOADS]: The skill uses MCP tools to fetch external online screenshots if a project URL is detected in the sitemap data, introducing a dependency on external sources controlled by the input data.
Audit Metadata