create-verification-skill
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill instructions require the agent to generate and then execute non-playtest checks ("run every non-playtest check it documents before delivery"). This constitutes execution of dynamically generated scripts or commands based on local project state.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where untrusted data from the repository (project files, instructions, scripts) is ingested and used to generate executable commands.\n
- Ingestion points: SKILL.md directs the agent to inspect repository instructions, project files, scripts, and test layouts.\n
- Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined for the processed project content.\n
- Capability inventory: The skill possesses the capability to execute shell commands as part of the verification process (e.g., rojo build, unit tests, integration tests).\n
- Sanitization: The instructions do not specify any sanitization or validation logic for the content extracted from the repository before it is used to generate or execute commands.\n- [COMMAND_EXECUTION]: The skill instructions specify the use of shell commands such as
rojo buildand various unit/integration test commands to verify the project state.
Audit Metadata