game-asset-production

Pass

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill defines a production pipeline for game sprites and animations, delegating execution to the fal-media MCP tool. It does not perform sensitive file access, network exfiltration, or privileged command execution. References to AI endpoints (fal-ai, openai) in assets/plan.json are consistent with the skill's stated purpose and use standard providers.
  • [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to read an external production plan guide (../genmedia/references/production-plans.md) and retrieve published guides via the fal-media tool. This creates a surface where instructions embedded in those external files could influence the agent's behavior. * Ingestion points: Local file reference in SKILL.md and dynamically retrieved guides via the fal-media tool. * Boundary markers: No delimiters or explicit instructions to ignore embedded commands are present. * Capability inventory: Media generation, image editing, and asset processing (matting, slicing). * Sanitization: No sanitization or validation of the external guide content is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 7, 2026, 07:17 AM