referral-program
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides strategic advice, benchmarks, and frameworks for referral marketing. It contains no executable scripts, shell commands, or external code dependencies. All external tools mentioned (such as Cello, Viral Loops, and ReferralCandy) are cited as informational examples of third-party services.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read local project context files (
.claude/project-context.mdor.cursor/project-context.md) to provide customized advice based on the user's product. While this involves ingesting external data, the skill possesses no capabilities (such as file writing, network exfiltration, or command execution) that could be exploited by malicious content within those files, rendering the attack surface benign.
Audit Metadata