skyline-wxss
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The main skill instructions in
SKILL.mddirect the agent to executenpx skyline-clicommands for style validation and listing supported CSS properties. - [EXTERNAL_DOWNLOADS]: The skill uses
npxto fetch theskyline-clipackage from the npm registry. This represents a standard developer workflow for the vendor's (wechat-miniprogram) own tooling. - [INDIRECT_PROMPT_INJECTION]: The skill implements a data ingestion surface by encouraging style checks on external project files.
- Ingestion points:
SKILL.mddemonstrates validation via stdin and file lists. - Boundary markers: The skill uses the
--jsonflag for output structuring but does not specify delimiters for the source WXSS content. - Capability inventory: Shell execution and file system access are leveraged for style verification.
- Sanitization: No explicit sanitization or filtering of style content is documented; the skill relies on the vendor tool's internal parsing logic.
Audit Metadata