studio-launch

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from various project-specific files to generate message variants and update logs, creating a surface for indirect prompt injection.
  • Ingestion points: Files including docs/LAUNCHES.md, docs/PLAN.md, productos/define/1-Product-Offer.md, productos/define/2-Customer-Persona.md, productos/define/3-Pricing-Strategy.md, productos/define/4-Mini-Launch.md, and various files in the productos/design/ directory.
  • Boundary markers: Absent. The skill does not define explicit delimiters to separate user content from system instructions when drafting messages.
  • Capability inventory: The agent drafts messages and writes to local files docs/LAUNCHES.md and productos/define/4-Mini-Launch.md.
  • Sanitization: Absent. The skill does not explicitly describe methods for sanitizing or validating input from these files.
  • [SAFE]: The skill functions as a procedural assistant for product management, with no evidence of malicious intent, unauthorized data access, or dangerous code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 10:33 PM
Security Audit — agent-trust-hub — studio-launch