design-gate
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a specific local script,
shared/scripts/gate_contract.py, to validate the gate results against a versioned record. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data such as Slice Contracts, design plans, and scope changes, which are then shared with other model contexts (specialists).
- Ingestion points: Workflow steps 1 and 3 explicitly identify plans, slice contracts, and changed scopes as inputs to the agent's context.
- Boundary markers: The instructions lack specific guidance on using delimiters or markers to separate untrusted content from system instructions.
- Capability inventory: The agent is authorized to execute a local Python script (
shared/scripts/gate_contract.py) based on the ingested content. - Sanitization: The workflow does not include steps for sanitizing or filtering instructions that might be embedded within the ingested design documents.
Audit Metadata