community-forum
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read project context from .claude/project-context.md or .cursor/project-context.md. Ingestion points: SKILL.md (Initial Assessment section). Boundary markers: None. Capability inventory: None (informational skill). Sanitization: None. While these files represent an ingestion point for external data, they are standard project files and the skill lacks capabilities to exploit this surface.
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. The content consists of legitimate marketing advice and engagement strategies for established community platforms.
Audit Metadata