coding-design-plan
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data including user-provided tasks, acceptance contracts, and existing project code.
- Ingestion points: Workflow step 1 instructs the agent to read external tasks, contracts, and codebases to propose a design shape.
- Boundary markers: No specific delimiters or instructions are provided to the agent to treat these external inputs strictly as data rather than instructions.
- Capability inventory: The skill is restricted to planning and returning structured text (Output Contract). It does not request or use capabilities for shell execution, network operations, or dynamic code evaluation.
- Sanitization: No sanitization or filtering of the ingested external content is specified in the instructions.
- [SAFE]: No signs of prompt injection, obfuscation, or unauthorized resource access were detected. The skill uses standard instructional language to define a software engineering workflow.
Audit Metadata