build-from-prd
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of Product Requirements Documents (PRDs) or specifications provided via connectors or pasted text. This creates a surface for indirect prompt injection where malicious instructions embedded in a PRD could attempt to override the agent's behavior or manipulate the design output.
- Ingestion points: The skill reads content from a 'connector to the requirements source' or 'pasted text' as specified in the Required Inputs section.
- Boundary markers: There are no explicit instructions or delimiters defined to isolate the PRD content from the agent's operational instructions, nor are there warnings to ignore embedded commands within the source text.
- Capability inventory: The skill possesses capabilities to modify the Figma environment, including creating frames, components, and instances, as well as altering prototype settings and layer structures.
- Sanitization: The workflow does not describe any sanitization, filtering, or validation steps for the incoming PRD text to mitigate potential injection attacks.
Audit Metadata