ljg-business-model
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as project descriptions, interview transcripts, and business plans, which can serve as a vector for indirect prompt injection attacks.\n
- Ingestion points: Data enters the system via user-provided text or documents as specified in
SKILL.mdandExample.json.\n - Boundary markers: The skill instructs the agent to categorize all findings as
given,inferred, orunknownand requires source references, which helps distinguish between provided data and potential malicious instructions.\n - Capability inventory: The skill utilizes a local TypeScript tool (
Tools/Render.ts) to transform data into HTML/PNG assets, involving file system write operations within the designated work directory.\n - Sanitization: The
Tools/Render.tsscript implements mandatory HTML entity encoding for all user-supplied text to prevent cross-site scripting (XSS) in the generated output cards.
Audit Metadata