stitch-design-system
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the Stitch API and HTML source code to generate CSS tokens and markdown documentation. This external data is used to populate templates without explicit sanitization.
- Ingestion points: Design metadata retrieved via the
get_projectandget_screentools, as well as HTML content downloaded from URLs provided by the API. - Boundary markers: The instructions do not define boundary markers or safety instructions to prevent the agent from obeying instructions that might be embedded in the design data.
- Capability inventory: The skill uses
Read,Write, andBashtools to manipulate files and execute scripts. - Sanitization: The skill maps API values (e.g.,
namedColors,fonts) directly into CSS and Tailwind artifacts without evidence of validation or escaping for the target formats. - [COMMAND_EXECUTION]: The skill utilizes the
Bashtool to execute a local script with arguments provided by a remote API. - Evidence: The command
bash scripts/fetch-stitch.sh "[htmlCode.downloadUrl]" "temp/source.html"inSKILL.mdexecutes a local script to download design assets using a remote URL as a parameter.
Audit Metadata