qa-game
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads external, user-provided data from files like
design-brief.mdand game source code to determine test logic and audit entity interactions. This data could contain malicious instructions that influence the agent's behavior during the test generation or auditing phase. - Ingestion points:
design-brief.md,package.json,vite.config.js,src/main.js,src/core/GameState.js,src/core/EventBus.js,src/core/Constants.js, and various scene files. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill when reading these files.
- Capability inventory: The skill has the capability to read and write to the local file system (including modifying
src/main.jsandpackage.json) and execute shell commands (npm test). - Sanitization: The skill does not specify any sanitization, validation, or filtering for the content ingested from the project files.
- [COMMAND_EXECUTION]: The skill performs shell operations to set up the testing environment and execute the generated test suite.
- Evidence: Utilizes
mkdir -p,cp(copying files from the plugin root), andnpm test/npm runcommands. - [DYNAMIC_EXECUTION]: The skill generates JavaScript and configuration files (
.js,.mjs) based on game logic templates and then executes them via a test runner. - Evidence: Creates test files in the
tests/directory and helpers inscripts/, which are subsequently executed during the verification step.
Audit Metadata