prototype
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to modify project configuration and task runner files (e.g.,
package.json,pyproject.toml,Makefile,justfile) to add execution scripts for prototypes. - [PROMPT_INJECTION]: The prototypes generated by this skill ingest external data (terminal keystrokes or URL search parameters), establishing a surface for indirect prompt injection if that data is subsequently used in agent reasoning steps.
- Ingestion points: Terminal input in logic prototypes (
LOGIC.md) and?variant=URL search parameters in UI prototypes (UI.md). - Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings for data ingested by the prototypes.
- Capability inventory: The skill facilitates modification of project build scripts and localized file system writes for state persistence ('scratch DBs').
- Sanitization: There are no explicit requirements for sanitizing or validating user-provided input before it is processed or rendered by the generated prototype code.
Audit Metadata