kai-launch
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project data, creating a potential surface for indirect prompt injection.
- Ingestion points: The skill automatically explores the codebase and reads files including
CLAUDE.md,README.md,PROJECT.md,package.json, and existing landing pages to extract product context. - Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions that might be embedded within the project files it scans.
- Capability inventory: The skill's primary capabilities involve file creation and modification within the
workspace/launch/directory and the project root. - Sanitization: No sanitization or validation logic is specified for the content extracted from the scanned project files before it is used to generate marketing assets.
Audit Metadata